[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-guyulongcs--Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising":3,"tool-guyulongcs--Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising":61},[4,18,26,36,44,53],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",147882,2,"2026-04-09T11:32:47",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",108111,"2026-04-08T11:23:26",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":32,"last_commit_at":50,"category_tags":51,"status":17},4721,"markitdown","microsoft\u002Fmarkitdown","MarkItDown 是一款由微软 AutoGen 团队打造的轻量级 Python 工具，专为将各类文件高效转换为 Markdown 格式而设计。它支持 PDF、Word、Excel、PPT、图片（含 OCR）、音频（含语音转录）、HTML 乃至 YouTube 链接等多种格式的解析，能够精准提取文档中的标题、列表、表格和链接等关键结构信息。\n\n在人工智能应用日益普及的今天，大语言模型（LLM）虽擅长处理文本，却难以直接读取复杂的二进制办公文档。MarkItDown 恰好解决了这一痛点，它将非结构化或半结构化的文件转化为模型“原生理解”且 Token 效率极高的 Markdown 格式，成为连接本地文件与 AI 分析 pipeline 的理想桥梁。此外，它还提供了 MCP（模型上下文协议）服务器，可无缝集成到 Claude Desktop 等 LLM 应用中。\n\n这款工具特别适合开发者、数据科学家及 AI 研究人员使用，尤其是那些需要构建文档检索增强生成（RAG）系统、进行批量文本分析或希望让 AI 助手直接“阅读”本地文件的用户。虽然生成的内容也具备一定可读性，但其核心优势在于为机器",93400,"2026-04-06T19:52:38",[52,14],"插件",{"id":54,"name":55,"github_repo":56,"description_zh":57,"stars":58,"difficulty_score":10,"last_commit_at":59,"category_tags":60,"status":17},4487,"LLMs-from-scratch","rasbt\u002FLLMs-from-scratch","LLMs-from-scratch 是一个基于 PyTorch 的开源教育项目，旨在引导用户从零开始一步步构建一个类似 ChatGPT 的大型语言模型（LLM）。它不仅是同名技术著作的官方代码库，更提供了一套完整的实践方案，涵盖模型开发、预训练及微调的全过程。\n\n该项目主要解决了大模型领域“黑盒化”的学习痛点。许多开发者虽能调用现成模型，却难以深入理解其内部架构与训练机制。通过亲手编写每一行核心代码，用户能够透彻掌握 Transformer 架构、注意力机制等关键原理，从而真正理解大模型是如何“思考”的。此外，项目还包含了加载大型预训练权重进行微调的代码，帮助用户将理论知识延伸至实际应用。\n\nLLMs-from-scratch 特别适合希望深入底层原理的 AI 开发者、研究人员以及计算机专业的学生。对于不满足于仅使用 API，而是渴望探究模型构建细节的技术人员而言，这是极佳的学习资源。其独特的技术亮点在于“循序渐进”的教学设计：将复杂的系统工程拆解为清晰的步骤，配合详细的图表与示例，让构建一个虽小但功能完备的大模型变得触手可及。无论你是想夯实理论基础，还是为未来研发更大规模的模型做准备",90106,"2026-04-06T11:19:32",[35,15,13,14],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":67,"readme_en":68,"readme_zh":69,"quickstart_zh":70,"use_case_zh":71,"hero_image_url":72,"owner_login":73,"owner_name":74,"owner_avatar_url":75,"owner_bio":76,"owner_company":77,"owner_location":78,"owner_email":79,"owner_twitter":80,"owner_website":81,"owner_url":82,"languages":83,"stars":88,"forks":89,"last_commit_at":90,"license":80,"difficulty_score":91,"env_os":92,"env_gpu":92,"env_ram":92,"env_deps":93,"category_tags":96,"github_topics":98,"view_count":32,"oss_zip_url":80,"oss_zip_packed_at":80,"status":17,"created_at":107,"updated_at":108,"faqs":109,"releases":110},5860,"guyulongcs\u002FAwesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising","Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising","Awesome Deep Learning papers for industrial Search, Recommendation and Advertisement. They focus on Embedding, Matching, Pre-Ranking, Ranking (CTR\u002FCVR prediction), Post Ranking, Relevance, LLM, Reinforcement Learning and so on.","Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising 是一份专为工业界打造的深度学习论文精选清单，聚焦于搜索、推荐系统与广告领域的核心算法演进。它系统性地梳理了从基础嵌入（Embedding）到复杂排序（Ranking）、重排序（Re-ranking），乃至大语言模型（LLM）与强化学习等前沿方向的关键研究成果。\n\n在海量数据场景下，如何高效提取特征、精准匹配用户意图并优化点击率（CTR）或转化率（CVR）是业界长期面临的挑战。这份资源通过分类整理经典与最新论文，帮助从业者快速定位技术脉络，避免在浩如烟海的文献中迷失方向。无论是重温 Word2vec、DeepWalk 等奠基之作，还是研究阿里巴巴、Pinterest 等大厂落地的 Billion-scale Embedding 与 PinSage 等实战方案，都能在此找到权威参考。\n\n该资源特别适合从事算法研发的工程师、攻读相关方向的研究生以及希望深入了解推荐机制的技术决策者。其独特亮点在于不仅涵盖学术界顶会（如 KDD、NIPS、ICLR）的理论突破，更","Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising 是一份专为工业界打造的深度学习论文精选清单，聚焦于搜索、推荐系统与广告领域的核心算法演进。它系统性地梳理了从基础嵌入（Embedding）到复杂排序（Ranking）、重排序（Re-ranking），乃至大语言模型（LLM）与强化学习等前沿方向的关键研究成果。\n\n在海量数据场景下，如何高效提取特征、精准匹配用户意图并优化点击率（CTR）或转化率（CVR）是业界长期面临的挑战。这份资源通过分类整理经典与最新论文，帮助从业者快速定位技术脉络，避免在浩如烟海的文献中迷失方向。无论是重温 Word2vec、DeepWalk 等奠基之作，还是研究阿里巴巴、Pinterest 等大厂落地的 Billion-scale Embedding 与 PinSage 等实战方案，都能在此找到权威参考。\n\n该资源特别适合从事算法研发的工程师、攻读相关方向的研究生以及希望深入了解推荐机制的技术决策者。其独特亮点在于不仅涵盖学术界顶会（如 KDD、NIPS、ICLR）的理论突破，更着重收录了具有大规模工业应用背景的实战论文，实现了理论与实践的紧密衔接。对于想要构建高效推荐系统或追踪行业技术风向的专业人士而言，这是一份极具价值的入门指南与进阶手册。","## Awesome Deep Learning papers for industrial Search, Recommendation and Advertisement. They focus on Embedding, Matching, Pre-Ranking, Ranking (CTR\u002FCVR prediction), Post Ranking, Relevance, LLM, Reinforcement Learning and so on.\n\n## 00_Embedding\n* [2013 (Google) (NIPS) [Word2vec] Distributed Representations of Words and Phrases and their Compositionality](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2013%20%28Google%29%20%28NIPS%29%20%5BWord2vec%5D%20Distributed%20Representations%20of%20Words%20and%20Phrases%20and%20their%20Compositionality.pdf) \u003Cbr \u002F>\n* [2014 (KDD) [DeepWalk]  DeepWalk - online learning of social representations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2014%20%28KDD%29%20%5BDeepWalk%5D%20%20DeepWalk%20-%20online%20learning%20of%20social%20representations.pdf) \u003Cbr \u002F>\n* [2015 (WWW) [LINE] LINE Large-scale Information Network Embedding](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2015%20%28WWW%29%20%5BLINE%5D%20LINE%20Large-scale%20Information%20Network%20Embedding.pdf) \u003Cbr \u002F>\n* [2016 (KDD) [Node2vec] node2vec - Scalable Feature Learning for Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2016%20%28KDD%29%20%5BNode2vec%5D%20node2vec%20-%20Scalable%20Feature%20Learning%20for%20Networks.pdf) \u003Cbr \u002F>\n* [2017 (ICLR) [GCN] Semi-supervised Classification with Graph Convolutional Networks ](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2017%20%28ICLR%29%20%5BGCN%5D%20Semi-supervised%20Classification%20with%20Graph%20Convolutional%20Networks%20.pdf) \u003Cbr \u002F>\n* [2017 (KDD) [Struc2vec] struc2vec - Learning Node Representations from Structural Identity](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2017%20%28KDD%29%20%5BStruc2vec%5D%20struc2vec%20-%20Learning%20Node%20Representations%20from%20Structural%20Identity.pdf) \u003Cbr \u002F>\n* [2017 (NIPS) [GraphSAGE] Inductive Representation Learning on Large Graphs](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2017%20%28NIPS%29%20%5BGraphSAGE%5D%20Inductive%20Representation%20Learning%20on%20Large%20Graphs.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (KDD) *[Alibaba Embedding] Billion-scale Commodity Embedding for E-commerce Recommendation in Alibaba](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2018%20%28Alibaba%29%20%28KDD%29%20%2A%5BAlibaba%20Embedding%5D%20Billion-scale%20Commodity%20Embedding%20for%20E-commerce%20Recommendation%20in%20Alibaba.pdf) \u003Cbr \u002F>\n* [2018 (ICLR) [GAT]  Graph Attention Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2018%20%28ICLR%29%20%5BGAT%5D%20%20Graph%20Attention%20Networks.pdf) \u003Cbr \u002F>\n* [2018 (Pinterest) (KDD) *[PinSage] Graph Convolutional Neural Networks for Web-Scale Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2018%20%28Pinterest%29%20%28KDD%29%20%2A%5BPinSage%5D%20Graph%20Convolutional%20Neural%20Networks%20for%20Web-Scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2018 (WSDM) [NetMF] Network embedding as matrix factorization - Unifying deepwalk, line, pte, and node2vec](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2018%20%28WSDM%29%20%5BNetMF%5D%20Network%20embedding%20as%20matrix%20factorization%20-%20Unifying%20deepwalk%2C%20line%2C%20pte%2C%20and%20node2vec.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) *[GATNE] Representation Learning for Attributed Multiplex Heterogeneous Network](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2019%20%28Alibaba%29%20%28KDD%29%20%2A%5BGATNE%5D%20Representation%20Learning%20for%20Attributed%20Multiplex%20Heterogeneous%20Network.pdf) \u003Cbr \u002F>\n\n## 01_Matching\n* [1994 (CSCW) [User-CF] GroupLens - An Open Architecture for Collaborative Filtering of Netnews](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F1994%20%28CSCW%29%20%5BUser-CF%5D%20GroupLens%20-%20An%20Open%20Architecture%20for%20Collaborative%20Filtering%20of%20Netnews.pdf) \u003Cbr \u002F>\n* [1998 (Microsoft) Empirical Analysis of Predictive Algorithms for Collaborative Filtering](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F1998%20%28Microsoft%29%20Empirical%20Analysis%20of%20Predictive%20Algorithms%20for%20Collaborative%20Filtering.pdf) \u003Cbr \u002F>\n* [2003 (Amazon) [Item-CF] Amazon.com recommendations - item-to-item collaborative filtering](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2003%20%28Amazon%29%20%5BItem-CF%5D%20Amazon.com%20recommendations%20-%20item-to-item%20collaborative%20filtering.pdf) \u003Cbr \u002F>\n* [2009 (Computer) [MF] Matrix factorization techniques for recommender systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2009%20%28Computer%29%20%5BMF%5D%20Matrix%20factorization%20techniques%20for%20recommender%20systems.pdf) \u003Cbr \u002F>\n* [2013 (Microsoft) (CIKM) [DSSM] Learning deep structured semantic models for web search using clickthrough data](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2013%20%28Microsoft%29%20%28CIKM%29%20%5BDSSM%5D%20Learning%20deep%20structured%20semantic%20models%20for%20web%20search%20using%20clickthrough%20data.pdf) \u003Cbr \u002F>\n* [2015 (KDD) [Sceptre] Inferring Networks of Substitutable and Complementary Products](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2015%20%28KDD%29%20%5BSceptre%5D%20Inferring%20Networks%20of%20Substitutable%20and%20Complementary%20Products.pdf) \u003Cbr \u002F>\n* [2016 (Google) (RecSys) **[Youtube DNN] Deep Neural Networks for YouTube Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2016%20%28Google%29%20%28RecSys%29%20%2A%2A%5BYoutube%20DNN%5D%20Deep%20Neural%20Networks%20for%20YouTube%20Recommendations.pdf) \u003Cbr \u002F>\n* [2018 (Airbnb) (KDD) *[Airbnb Embedding] Real-time Personalization using Embeddings for Search Ranking at Airbnb](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2018%20%28Airbnb%29%20%28KDD%29%20%2A%5BAirbnb%20Embedding%5D%20Real-time%20Personalization%20using%20Embeddings%20for%20Search%20Ranking%20at%20Airbnb.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (KDD) * [TDM] Learning Tree-based Deep Model for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2018%20%28Alibaba%29%20%28KDD%29%20%2A%20%5BTDM%5D%20Learning%20Tree-based%20Deep%20Model%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2018 (Pinterest) (KDD) *[PinSage] Graph Convolutional Neural Networks for Web-Scale Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2018%20%28Pinterest%29%20%28KDD%29%20%2A%5BPinSage%5D%20Graph%20Convolutional%20Neural%20Networks%20for%20Web-Scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (CIKM) **[MIND] Multi-Interest Network with Dynamic Routing for Recommendation at Tmall](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Alibaba%29%20%28CIKM%29%20%2A%2A%5BMIND%5D%20Multi-Interest%20Network%20with%20Dynamic%20Routing%20for%20Recommendation%20at%20Tmall.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (CIKM) *[SDM] SDM - Sequential deep matching model for online large-scale recommender system](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Alibaba%29%20%28CIKM%29%20%2A%5BSDM%5D%20SDM%20-%20Sequential%20deep%20matching%20model%20for%20online%20large-scale%20recommender%20system.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (NIPS) *[JTM] Joint Optimization of Tree-based Index and Deep Model for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Alibaba%29%20%28NIPS%29%20%2A%5BJTM%5D%20Joint%20Optimization%20of%20Tree-based%20Index%20and%20Deep%20Model%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2019 (Amazon) (KDD) Semantic Product Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Amazon%29%20%28KDD%29%20Semantic%20Product%20Search.pdf) \u003Cbr \u002F>\n* [2019 (Baidu) (KDD) *[MOBIUS] MOBIUS - Towards the Next Generation of Query-Ad Matching in Baidu's Sponsored Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Baidu%29%20%28KDD%29%20%2A%5BMOBIUS%5D%20MOBIUS%20-%20Towards%20the%20Next%20Generation%20of%20Query-Ad%20Matching%20in%20Baidu%27s%20Sponsored%20Search.pdf) \u003Cbr \u002F>\n* [2019 (Google) (RecSys) **[Two-Tower] Sampling-Bias-Corrected Neural Modeling for Large Corpus Item Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Google%29%20%28RecSys%29%20%2A%2A%5BTwo-Tower%5D%20Sampling-Bias-Corrected%20Neural%20Modeling%20for%20Large%20Corpus%20Item%20Recommendations.pdf) \u003Cbr \u002F>\n* [2019 (Google) (WSDM) *[Top-K Off-Policy] Top-K Off-Policy Correction for a REINFORCE Recommender System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Google%29%20%28WSDM%29%20%2A%5BTop-K%20Off-Policy%5D%20Top-K%20Off-Policy%20Correction%20for%20a%20REINFORCE%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2019 [Tencent] (KDD) A User-Centered Concept Mining System for Query and Document Understanding at Tencent](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%5BTencent%5D%20%28KDD%29%20A%20User-Centered%20Concept%20Mining%20System%20for%20Query%20and%20Document%20Understanding%20at%20Tencent.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (Arxiv) [SWING] Large Scale Product Graph Construction for Recommendation in E-commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Alibaba%29%20%28Arxiv%29%20%5BSWING%5D%20Large%20Scale%20Product%20Graph%20Construction%20for%20Recommendation%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (ICML) [OTM] Learning Optimal Tree Models under Beam Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Alibaba%29%20%28ICML%29%20%5BOTM%5D%20Learning%20Optimal%20Tree%20Models%20under%20Beam%20Search.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (KDD) *[ComiRec] Controllable Multi-Interest Framework for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Alibaba%29%20%28KDD%29%20%2A%5BComiRec%5D%20Controllable%20Multi-Interest%20Framework%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2020 (Facebook) (KDD) **[Embedding for Facebook Search] Embedding-based Retrieval in Facebook Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Facebook%29%20%28KDD%29%20%2A%2A%5BEmbedding%20for%20Facebook%20Search%5D%20Embedding-based%20Retrieval%20in%20Facebook%20Search.pdf) \u003Cbr \u002F>\n* [2020 (Google) (WWW) *[MNS] Mixed Negative Sampling for Learning Two-tower Neural Networks in Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Google%29%20%28WWW%29%20%2A%5BMNS%5D%20Mixed%20Negative%20Sampling%20for%20Learning%20Two-tower%20Neural%20Networks%20in%20Recommendations.pdf) \u003Cbr \u002F>\n* [2020 (JD) (CIKM) *[DecGCN] Decoupled Graph Convolution Network for Inferring Substitutable and Complementary Items](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28JD%29%20%28CIKM%29%20%2A%5BDecGCN%5D%20Decoupled%20Graph%20Convolution%20Network%20for%20Inferring%20Substitutable%20and%20Complementary%20Items.pdf) \u003Cbr \u002F>\n* [2020 (JD) (SIGIR) [DPSR] Towards Personalized and Semantic Retrieval - An End-to-EndSolution for E-commerce Search via Embedding Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28JD%29%20%28SIGIR%29%20%5BDPSR%5D%20Towards%20Personalized%20and%20Semantic%20Retrieval%20-%20An%20End-to-EndSolution%20for%20E-commerce%20Search%20via%20Embedding%20Learning.pdf) \u003Cbr \u002F>\n* [2020 (Microsoft) (Arxiv) TwinBERT - Distilling Knowledge to Twin-Structured BERT Models for Efficient Retrieval](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Microsoft%29%20%28Arxiv%29%20TwinBERT%20-%20Distilling%20Knowledge%20to%20Twin-Structured%20BERT%20Models%20for%20Efficient%20Retrieval.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (KDD) *  [MGDSPR] Embedding-based Product Retrieval in Taobao Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Alibaba%29%20%28KDD%29%20%2A%20%20%5BMGDSPR%5D%20Embedding-based%20Product%20Retrieval%20in%20Taobao%20Search.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (SIGIR) * [PDN] Path-based Deep Network for Candidate Item Matching in Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Alibaba%29%20%28SIGIR%29%20%2A%20%5BPDN%5D%20Path-based%20Deep%20Network%20for%20Candidate%20Item%20Matching%20in%20Recommenders.pdf) \u003Cbr \u002F>\n* [2021 (Amazon) (KDD) Extreme Multi-label Learning for Semantic Matching in Product Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Amazon%29%20%28KDD%29%20Extreme%20Multi-label%20Learning%20for%20Semantic%20Matching%20in%20Product%20Search.pdf) \u003Cbr \u002F>\n* [2021 (Baidu) (KDD) Pre-trained Language Model for Web-scale Retrieval in Baidu Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Baidu%29%20%28KDD%29%20Pre-trained%20Language%20Model%20for%20Web-scale%20Retrieval%20in%20Baidu%20Search.pdf) \u003Cbr \u002F>\n* [2021 (Bytedance) (Arxiv) [DR] Deep Retrieval - Learning A Retrievable Structure for Large-Scale Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Bytedance%29%20%28Arxiv%29%20%5BDR%5D%20Deep%20Retrieval%20-%20Learning%20A%20Retrievable%20Structure%20for%20Large-Scale%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021 (Meituan) (DLP-KDD) [DAT]A Dual Augmented Two-tower Model for Online Large-scale Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Meituan%29%20%28DLP-KDD%29%20%5BDAT%5DA%20Dual%20Augmented%20Two-tower%20Model%20for%20Online%20Large-scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (CIKM) **[NANN] Approximate Nearest Neighbor Search under Neural Similarity Metric for Large-Scale Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2022%20%28Alibaba%29%20%28CIKM%29%20%2A%2A%5BNANN%5D%20Approximate%20Nearest%20Neighbor%20Search%20under%20Neural%20Similarity%20Metric%20for%20Large-Scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (CIKM) [CLE-QR] Query Rewriting in TaoBao Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2022%20%28Alibaba%29%20%28CIKM%29%20%5BCLE-QR%5D%20Query%20Rewriting%20in%20TaoBao%20Search.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) **(CIKM) [MOPPR] Multi-Objective Personalized Product Retrieval in Taobao Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2022%20%28Alibaba%29%20%2A%2A%28CIKM%29%20%5BMOPPR%5D%20Multi-Objective%20Personalized%20Product%20Retrieval%20in%20Taobao%20Search.pdf) \u003Cbr \u002F>\n* [2024 (Bytedance) (KDD) [Trinity] Trinity - Syncretizing Multi-:Long-Tail:Long-Term Interests All in One](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2024%20%28Bytedance%29%20%28KDD%29%20%5BTrinity%5D%20Trinity%20-%20Syncretizing%20Multi-%3ALong-Tail%3ALong-Term%20Interests%20All%20in%20One.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) ** [GR] Actions Speak Louder than Words - Trillion-Parameter Sequential Transducers for Generative Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2024%20%28Meta%29%20%28Arxiv%29%20%2A%2A%20%5BGR%5D%20Actions%20Speak%20Louder%20than%20Words%20-%20Trillion-Parameter%20Sequential%20Transducers%20for%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) (Arxiv) [LongRetriever] LongRetriever - Towards Ultra-Long Sequence based Candidate Retrieval for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Bytedance%29%20%28Arxiv%29%20%5BLongRetriever%5D%20LongRetriever%20-%20Towards%20Ultra-Long%20Sequence%20based%20Candidate%20Retrieval%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) (KDD) [VQ] Real-time Indexing for Large-scale Recommendation by Streaming Vector Quantization Retriever](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Bytedance%29%20%28KDD%29%20%5BVQ%5D%20Real-time%20Indexing%20for%20Large-scale%20Recommendation%20by%20Streaming%20Vector%20Quantization%20Retriever.pdf) \u003Cbr \u002F>\n* [2025 (JD) (KDD) [UniERF] UniERF - A Uniform Embedding-based Retrieval Framework for E-Commerce Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28JD%29%20%28KDD%29%20%5BUniERF%5D%20UniERF%20-%20A%20Uniform%20Embedding-based%20Retrieval%20Framework%20for%20E-Commerce%20Search.pdf) \u003Cbr \u002F>\n* [2025 (Meta) (KDD) [MTMH] Optimizing Recall or Relevance? A Multi-Task Multi-Head Approach for Item-to-Item Retrieval in Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Meta%29%20%28KDD%29%20%5BMTMH%5D%20Optimizing%20Recall%20or%20Relevance%3F%20A%20Multi-Task%20Multi-Head%20Approach%20for%20Item-to-Item%20Retrieval%20in%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Meta) [RADAR ] RADAR - Recall Augmentation through Deferred Asynchronous Retrieval](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Meta%29%20%5BRADAR%20%5D%20RADAR%20-%20Recall%20Augmentation%20through%20Deferred%20Asynchronous%20Retrieval.pdf) \u003Cbr \u002F>\n* [2025 (Tencent) (Arxiv) An Efficient Embedding Based Ad Retrieval with GPU-Powered Feature Interaction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Tencent%29%20%28Arxiv%29%20An%20Efficient%20Embedding%20Based%20Ad%20Retrieval%20with%20GPU-Powered%20Feature%20Interaction.pdf) \u003Cbr \u002F>\n\n#### ANN\n* [2017 (Arxiv) (Meta) [FAISS] Billion-scale similarity search with GPUs](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FANN\u002F2017%20%28Arxiv%29%20%28Meta%29%20%5BFAISS%5D%20Billion-scale%20similarity%20search%20with%20GPUs.pdf) \u003Cbr \u002F>\n* [2020 (PAMI) [HNSW] Efficient and Robust Approximate Nearest Neighbor Search Using Hierarchical Navigable Small World Graphs](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FANN\u002F2020%20%28PAMI%29%20%5BHNSW%5D%20Efficient%20and%20Robust%20Approximate%20Nearest%20Neighbor%20Search%20Using%20Hierarchical%20Navigable%20Small%20World%20Graphs.pdf) \u003Cbr \u002F>\n* [2021 (TPAMI) [IVF-PQ] Product Quantization for Nearest Neighbor Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FANN\u002F2021%20%28TPAMI%29%20%5BIVF-PQ%5D%20Product%20Quantization%20for%20Nearest%20Neighbor%20Search.pdf) \u003Cbr \u002F>\n\n#### Graph_Neural_Networks\n* [2017 (ICLR) [GCN] Semi-Supervised Classification with Graph Convolutional Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2017%20%28ICLR%29%20%5BGCN%5D%20Semi-Supervised%20Classification%20with%20Graph%20Convolutional%20Networks.pdf) \u003Cbr \u002F>\n* [2018 (ICLR) [GAT] Graph Attention Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2018%20%28ICLR%29%20%5BGAT%5D%20Graph%20Attention%20Networks.pdf) \u003Cbr \u002F>\n* [2018 (Pinterest) (KDD) [PinSage] Graph Convolutional Neural Networks for Web-Scale Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2018%20%28Pinterest%29%20%28KDD%29%20%5BPinSage%5D%20Graph%20Convolutional%20Neural%20Networks%20for%20Web-Scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) [IntentGC] IntentGC - a Scalable Graph Convolution Framework Fusing Heterogeneous Information for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BIntentGC%5D%20IntentGC%20-%20a%20Scalable%20Graph%20Convolution%20Framework%20Fusing%20Heterogeneous%20Information%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) [MEIRec] Metapath-guided Heterogeneous Graph Neural Network for Intent Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BMEIRec%5D%20Metapath-guided%20Heterogeneous%20Graph%20Neural%20Network%20for%20Intent%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (SIGIR) [GIN] Graph Intention Network for Click-through Rate Prediction in Sponsored Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2019%20%28Alibaba%29%20%28SIGIR%29%20%5BGIN%5D%20Graph%20Intention%20Network%20for%20Click-through%20Rate%20Prediction%20in%20Sponsored%20Search.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (SIGIR) [ATBRG] ATBRG - Adaptive Target-Behavior Relational Graph Network for Effective Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2020%20%28Alibaba%29%20%28SIGIR%29%20%5BATBRG%5D%20ATBRG%20-%20Adaptive%20Target-Behavior%20Relational%20Graph%20Network%20for%20Effective%20Recommendation.pdf) \u003Cbr \u002F>\n\n#### LLM_Matching\n* [2021 (Baidu) (KDD) Pre-trained Language Model for Web-scale Retrieval in Baidu Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2021%20%28Baidu%29%20%28KDD%29%20Pre-trained%20Language%20Model%20for%20Web-scale%20Retrieval%20in%20Baidu%20Search.pdf) \u003Cbr \u002F>\n* [2023 (Google) (NIPS) [TIGER] Recommender Systems with Generative Retrieval](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2023%20%28Google%29%20%28NIPS%29%20%5BTIGER%5D%20Recommender%20Systems%20with%20Generative%20Retrieval.pdf) \u003Cbr \u002F>\n* [2024 (Alibaba) (WWW) [BEQUE] Large Language Model based Long-tail Query Rewriting in Taobao Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Alibaba%29%20%28WWW%29%20%5BBEQUE%5D%20Large%20Language%20Model%20based%20Long-tail%20Query%20Rewriting%20in%20Taobao%20Search.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (Arxiv) [KuaiFormer] KuaiFormer - Transformer-Based Retrieval at Kuaishou](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Kuaishou%29%20%28Arxiv%29%20%5BKuaiFormer%5D%20KuaiFormer%20-%20Transformer-Based%20Retrieval%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) ** [GR] Actions Speak Louder than Words - Trillion-Parameter Sequential Transducers for Generative Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Meta%29%20%28Arxiv%29%20%2A%2A%20%5BGR%5D%20Actions%20Speak%20Louder%20than%20Words%20-%20Trillion-Parameter%20Sequential%20Transducers%20for%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) Unifying Generative and Dense Retrieval for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Meta%29%20%28Arxiv%29%20Unifying%20Generative%20and%20Dense%20Retrieval%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 (Xiaohongshu) (WWW) [NoteLLM] NoteLLM - Multimodal Large Representation Models for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Xiaohongshu%29%20%28WWW%29%20%5BNoteLLM%5D%20NoteLLM%20-%20Multimodal%20Large%20Representation%20Models%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025  (Kuaishou) (Arxiv)[OneRec] OneRec - Unifying Retrieve and Rank with Generative Recommender and Preference Alignment](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%20%28Kuaishou%29%20%28Arxiv%29%5BOneRec%5D%20OneRec%20-%20Unifying%20Retrieve%20and%20Rank%20with%20Generative%20Recommender%20and%20Preference%20Alignment.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [TBGRecall] TBGRecall - A Generative Retrieval Model for E-commerce Recommendation Scenarios](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BTBGRecall%5D%20TBGRecall%20-%20A%20Generative%20Retrieval%20Model%20for%20E-commerce%20Recommendation%20Scenarios.pdf) \u003Cbr \u002F>\n* [2025 (Arxiv) (Tencent) [RARE] Real-time Ad retrieval via LLM-generative Commercial Intention for Sponsored Search Advertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Arxiv%29%20%28Tencent%29%20%5BRARE%5D%20Real-time%20Ad%20retrieval%20via%20LLM-generative%20Commercial%20Intention%20for%20Sponsored%20Search%20Advertising.pdf) \u003Cbr \u002F>\n* [2025 (Baidu) (Arxiv) [COBRA] Sparse Meets Dense -Unified Generative Recommendations with Cascaded Sparse-Dense Representations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Baidu%29%20%28Arxiv%29%20%5BCOBRA%5D%20Sparse%20Meets%20Dense%20-Unified%20Generative%20Recommendations%20with%20Cascaded%20Sparse-Dense%20Representations.pdf) \u003Cbr \u002F>\n* [2025 (Google) [PLUM] PLUM - Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Google%29%20%5BPLUM%5D%20PLUM%20-%20Adapting%20Pre-trained%20Language%20Models%20for%20Industrial-scale%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2025 (JD) (Arxiv) [GRAM] Generative Retrieval and Alignment Model - A New Paradigm for E-commerce Retrieval](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28JD%29%20%28Arxiv%29%20%5BGRAM%5D%20Generative%20Retrieval%20and%20Alignment%20Model%20-%20A%20New%20Paradigm%20for%20E-commerce%20Retrieval.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (AAAI) [Align3GR] Align3GR - Unified Multi-Level Alignment for LLM-based Generative Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Kuaishou%29%20%28AAAI%29%20%5BAlign3GR%5D%20Align3GR%20-%20Unified%20Multi-Level%20Alignment%20for%20LLM-based%20Generative%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [LARM] LLM-Alignment Live-Streaming Recommendationpdf](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BLARM%5D%20LLM-Alignment%20Live-Streaming%20Recommendationpdf.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [LEARN] LEARN - Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Application](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BLEARN%5D%20LEARN%20-%20Knowledge%20Adaptation%20from%20Large%20Language%20Model%20to%20Recommendation%20for%20Practical%20Industrial%20Application.pdf) \u003Cbr \u002F>\n* [2025 (Meta) (Arxiv) [DRAMA] DRAMA - Diverse Augmentation from Large Language Models to Smaller Dense Retrievers](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Meta%29%20%28Arxiv%29%20%5BDRAMA%5D%20DRAMA%20-%20Diverse%20Augmentation%20from%20Large%20Language%20Models%20to%20Smaller%20Dense%20Retrievers.pdf) \u003Cbr \u002F>\n* [2025 (Meta) (Arxiv) [ROO] Request-Only Optimization for Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Meta%29%20%28Arxiv%29%20%5BROO%5D%20Request-Only%20Optimization%20for%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Pinterest) [PinRec] PinRec - Outcome-Conditioned, Multi-Token Generative Retrieval for Industry-Scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Pinterest%29%20%5BPinRec%5D%20PinRec%20-%20Outcome-Conditioned%2C%20Multi-Token%20Generative%20Retrieval%20for%20Industry-Scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Xiaohongshu) （KDD) [NoteLLM-2] NoteLLM-2 - Multimodal Large Representation Models for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Xiaohongshu%29%20%EF%BC%88KDD%29%20%5BNoteLLM-2%5D%20NoteLLM-2%20-%20Multimodal%20Large%20Representation%20Models%20for%20Recommendation.pdf) \u003Cbr \u002F>\n\n## 02_Pre-ranking\n* [2020 (Alibaba) (DLP-KDD) [COLD] COLD - Towards the Next Generation of Pre-Ranking System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2020%20%28Alibaba%29%20%28DLP-KDD%29%20%5BCOLD%5D%20COLD%20-%20Towards%20the%20Next%20Generation%20of%20Pre-Ranking%20System.pdf) \u003Cbr \u002F>\n* [2022 (Huawei) (SIGIR) [RankFlow] RankFlow - JointOptimization ofMulti-Stage CascadeRanking SystemsasFlows](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2022%20%28Huawei%29%20%28SIGIR%29%20%5BRankFlow%5D%20RankFlow%20-%20JointOptimization%20ofMulti-Stage%20CascadeRanking%20SystemsasFlows.pdf) \u003Cbr \u002F>\n* [2022 （Huawei) (CIKM) [IntTower] IntTower - the Next Generation of Two-Tower Model for Pre-Ranking System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2022%20%EF%BC%88Huawei%29%20%28CIKM%29%20%5BIntTower%5D%20IntTower%20-%20the%20Next%20Generation%20of%20Two-Tower%20Model%20for%20Pre-Ranking%20System.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (CIKM) [COPR] COPR - Consistency-Oriented Pre-Ranking for Online Advertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BCOPR%5D%20COPR%20-%20Consistency-Oriented%20Pre-Ranking%20for%20Online%20Advertising.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (KDD) [ASMOL] Rethinking the Role of Pre-ranking in Large-scale E-Commerce Searching System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2023%20%28Alibaba%29%20%28KDD%29%20%5BASMOL%5D%20Rethinking%20the%20Role%20of%20Pre-ranking%20in%20Large-scale%20E-Commerce%20Searching%20System.pdf) \u003Cbr \u002F>\n* [2025 (Tencent) (Arxiv) [HIT] HIT Model - A Hierarchical Interaction-Enhanced Two-Tower Model for Pre-Ranking Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2025%20%28Tencent%29%20%28Arxiv%29%20%5BHIT%5D%20HIT%20Model%20-%20A%20Hierarchical%20Interaction-Enhanced%20Two-Tower%20Model%20for%20Pre-Ranking%20Systems.pdf) \u003Cbr \u002F>\n\n## 03_Ranking\n* [2014 (ADKDD) (Facebook) Practical Lessons from Predicting Clicks on Ads at Facebook](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2014%20%28ADKDD%29%20%28Facebook%29%20Practical%20Lessons%20from%20Predicting%20Clicks%20on%20Ads%20at%20Facebook.pdf) \u003Cbr \u002F>\n* [2016 (Google) (DLRS) **[Wide & Deep] Wide & Deep Learning for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2016%20%28Google%29%20%28DLRS%29%20%2A%2A%5BWide%20%26%20Deep%5D%20Wide%20%26%20Deep%20Learning%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2016 (Google) (RecSys) **[Youtube DNN] Deep Neural Networks for YouTube Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2016%20%28Google%29%20%28RecSys%29%20%2A%2A%5BYoutube%20DNN%5D%20Deep%20Neural%20Networks%20for%20YouTube%20Recommendations.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (KDD) **[DIN] Deep Interest Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2018%20%28Alibaba%29%20%28KDD%29%20%2A%2A%5BDIN%5D%20Deep%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (AAAI) **[DIEN] Deep Interest Evolution Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2019%20%28Alibaba%29%20%28AAAI%29%20%2A%2A%5BDIEN%5D%20Deep%20Interest%20Evolution%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (CIKM) ** [AutoInt] AutoInt -Automatic Feature Interaction Learning via Self-Attentive Neural Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2019%20%28CIKM%29%20%2A%2A%20%5BAutoInt%5D%20AutoInt%20-Automatic%20Feature%20Interaction%20Learning%20via%20Self-Attentive%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2019 (Facebook) (Arxiv) [DLRM] (Facebook) Deep Learning Recommendation Model for Personalization and Recommendation Systems, Facebook](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2019%20%28Facebook%29%20%28Arxiv%29%20%5BDLRM%5D%20%28Facebook%29%20Deep%20Learning%20Recommendation%20Model%20for%20Personalization%20and%20Recommendation%20Systems%2C%20Facebook.pdf) \u003Cbr \u002F>\n* [2019 (Google) (Recsys) ** [Youtube Multi-task] Recommending what video to watch next - a multitask ranking system](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2019%20%28Google%29%20%28Recsys%29%20%2A%2A%20%5BYoutube%20Multi-task%5D%20Recommending%20what%20video%20to%20watch%20next%20-%20a%20multitask%20ranking%20system.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (Arxiv) ** [SIM] Search-based User Interest Modeling with Lifelong Sequential Behavior Data for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2020%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BSIM%5D%20Search-based%20User%20Interest%20Modeling%20with%20Lifelong%20Sequential%20Behavior%20Data%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (NIPS) Neuron-level Structured Pruning using Polarization Regularizer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2020%20%28Alibaba%29%20%28NIPS%29%20Neuron-level%20Structured%20Pruning%20using%20Polarization%20Regularizer.pdf) \u003Cbr \u002F>\n* [2020 (JD) (CIKM) **[DMT] Deep Multifaceted Transformers for Multi-objective Ranking in Large-Scale E-commerce Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2020%20%28JD%29%20%28CIKM%29%20%2A%2A%5BDMT%5D%20Deep%20Multifaceted%20Transformers%20for%20Multi-objective%20Ranking%20in%20Large-Scale%20E-commerce%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2020 (Tencent) (Recsys) **  [PLE] Progressive Layered Extraction (PLE) - A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2020%20%28Tencent%29%20%28Recsys%29%20%2A%2A%20%20%5BPLE%5D%20Progressive%20Layered%20Extraction%20%28PLE%29%20-%20A%20Novel%20Multi-Task%20Learning%20%28MTL%29%20Model%20for%20Personalized%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (CIKM) * [ZEUS] Self-Supervised Learning on Users’ Spontaneous Behaviors for Multi-Scenario Ranking in E-commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%2A%20%5BZEUS%5D%20Self-Supervised%20Learning%20on%20Users%E2%80%99%20Spontaneous%20Behaviors%20for%20Multi-Scenario%20Ranking%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (CIKM) [STAR] One Model to Serve All - Star Topology Adaptive Recommender for Multi-Domain CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%5BSTAR%5D%20One%20Model%20to%20Serve%20All%20-%20Star%20Topology%20Adaptive%20Recommender%20for%20Multi-Domain%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Google) (WWW) * [DCN V2] DCN V2 - Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2021%20%28Google%29%20%28WWW%29%20%2A%20%5BDCN%20V2%5D%20DCN%20V2%20-%20Improved%20Deep%20%26%20Cross%20Network%20and%20Practical%20Lessons%20for%20Web-scale%20Learning%20to%20Rank%20Systems.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (Arxiv) ** [ETA] Efficient Long Sequential User Data Modeling for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2022%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BETA%5D%20Efficient%20Long%20Sequential%20User%20Data%20Modeling%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (WSDM) Modeling Users’ Contextualized Page-wise Feedback for Click-Through Rate Prediction in E-commerce Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2022%20%28Alibaba%29%20%28WSDM%29%20Modeling%20Users%E2%80%99%20Contextualized%20Page-wise%20Feedback%20for%20Click-Through%20Rate%20Prediction%20in%20E-commerce%20Search.pdf) \u003Cbr \u002F>\n* [2022 (Meta) ** (Arxiv) DHEN - A Deep and Hierarchical Ensemble Network for Large-Scale Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2022%20%28Meta%29%20%2A%2A%20%28Arxiv%29%20DHEN%20-%20A%20Deep%20and%20Hierarchical%20Ensemble%20Network%20for%20Large-Scale%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (Arxiv) [ESLM] Entire Space Learning Framework - Unbias Conversion Rate Prediction in Full Stages of Recommender System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Alibaba%29%20%28Arxiv%29%20%5BESLM%5D%20Entire%20Space%20Learning%20Framework%20-%20Unbias%20Conversion%20Rate%20Prediction%20in%20Full%20Stages%20of%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2023 (Google) (Arxiv) On the Factory Floor - ML Engineering for Industrial-Scale Ads Recommendation Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Google%29%20%28Arxiv%29%20On%20the%20Factory%20Floor%20-%20ML%20Engineering%20for%20Industrial-Scale%20Ads%20Recommendation%20Models.pdf) \u003Cbr \u002F>\n* [2023 (Google) ** (Arxiv) [Hiformer] Hiformer - Heterogeneous Feature Interactions Learning with Transformers for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Google%29%20%2A%2A%20%28Arxiv%29%20%5BHiformer%5D%20Hiformer%20-%20Heterogeneous%20Feature%20Interactions%20Learning%20with%20Transformers%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2023 (Kuaishou) (Arixiv) [TWIN] TWIN - TWo-stage Interest Network for Lifelong User Behavior Modeling in CTR Prediction at Kuaishou](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Kuaishou%29%20%28Arixiv%29%20%5BTWIN%5D%20TWIN%20-%20TWo-stage%20Interest%20Network%20for%20Lifelong%20User%20Behavior%20Modeling%20in%20CTR%20Prediction%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2023 (Kuaishou) (KDD) [PEPNet] PEPNet - Parameter and Embedding Personalized Network for Infusing with Personalized Prior Information](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Kuaishou%29%20%28KDD%29%20%5BPEPNet%5D%20PEPNet%20-%20Parameter%20and%20Embedding%20Personalized%20Network%20for%20Infusing%20with%20Personalized%20Prior%20Information.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (CIKM) [TWINv2] TWIN V2 - Scaling Ultra-Long User Behavior Sequence Modeling for Enhanced CTR Prediction at Kuaishou](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2024%20%28Kuaishou%29%20%28CIKM%29%20%5BTWINv2%5D%20TWIN%20V2%20-%20Scaling%20Ultra-Long%20User%20Behavior%20Sequence%20Modeling%20for%20Enhanced%20CTR%20Prediction%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) ** [GR] Actions Speak Louder than Words - Trillion-Parameter Sequential Transducers for Generative Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2024%20%28Meta%29%20%28Arxiv%29%20%2A%2A%20%5BGR%5D%20Actions%20Speak%20Louder%20than%20Words%20-%20Trillion-Parameter%20Sequential%20Transducers%20for%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024 (Meta) ** (PMLR) [Wukong] Wukong - Towards a Scaling Law for Large-Scale Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2024%20%28Meta%29%20%2A%2A%20%28PMLR%29%20%5BWukong%5D%20Wukong%20-%20Towards%20a%20Scaling%20Law%20for%20Large-Scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (Arxiv) [LONGER] LONGER - Scaling Up Long Sequence Modeling in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BLONGER%5D%20LONGER%20-%20Scaling%20Up%20Long%20Sequence%20Modeling%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (Arxiv) [RankMixer] RankMixer - Scaling Up Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 （Bytedance) ** (Arxiv) [STCA] Make It Long, Keep It Fast - End-to-End 10k-Sequence Modeling at Billion Scale on Douyin](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2025%20%EF%BC%88Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BSTCA%5D%20Make%20It%20Long%2C%20Keep%20It%20Fast%20-%20End-to-End%2010k-Sequence%20Modeling%20at%20Billion%20Scale%20on%20Douyin.pdf) \u003Cbr \u002F>\n* [2026 (Meta) (KDD) [Lattice] Meta Lattice - Model Space Redesign for Cost-Effective Industry-Scale Ads Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2026%20%28Meta%29%20%28KDD%29%20%5BLattice%5D%20Meta%20Lattice%20-%20Model%20Space%20Redesign%20for%20Cost-Effective%20Industry-Scale%20Ads%20Recommendations.pdf) \u003Cbr \u002F>\n\n#### Activation-Function\n* [2020(Arxiv)  [GLU] GLU Variants Improve Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FActivation-Function\u002F2020%28Arxiv%29%20%20%5BGLU%5D%20GLU%20Variants%20Improve%20Transformer.pdf) \u003Cbr \u002F>\n\n#### Calibration\n* [2014 (ADKDD) (Facebook) Practical Lessons from Predicting Clicks on Ads at Facebook](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FCalibration\u002F2014%20%28ADKDD%29%20%28Facebook%29%20Practical%20Lessons%20from%20Predicting%20Clicks%20on%20Ads%20at%20Facebook.pdf) \u003Cbr \u002F>\n* [2014 (TIST) Simple and scalable response prediction for display advertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FCalibration\u002F2014%20%28TIST%29%20Simple%20and%20scalable%20response%20prediction%20for%20display%20advertising.pdf) \u003Cbr \u002F>\n* [2023 Classifier Calibration with ROC-Regularized Isotonic Regression](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FCalibration\u002F2023%20Classifier%20Calibration%20with%20ROC-Regularized%20Isotonic%20Regression.pdf) \u003Cbr \u002F>\n\n#### Classic\n* [2016 (ICLR) [GRU4Rec] Session-based Recommendations with Recurrent Neural Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FClassic\u002F2016%20%28ICLR%29%20%5BGRU4Rec%5D%20Session-based%20Recommendations%20with%20Recurrent%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2017 (Amazon) (IEEE) Two decades of recommender systems at Amazon.com](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FClassic\u002F2017%20%28Amazon%29%20%28IEEE%29%20Two%20decades%20of%20recommender%20systems%20at%20Amazon.com.pdf) \u003Cbr \u002F>\n\n#### DNN\n* [2019 (KDD) (Airbnb) Applying Deep Learning To Airbnb Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDNN\u002F2019%20%28KDD%29%20%28Airbnb%29%20Applying%20Deep%20Learning%20To%20Airbnb%20Search.pdf) \u003Cbr \u002F>\n* [2020 (Airbnb) (KDD) Improving Deep Learning For Airbnb Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDNN\u002F2020%20%28Airbnb%29%20%28KDD%29%20Improving%20Deep%20Learning%20For%20Airbnb%20Search.pdf) \u003Cbr \u002F>\n\n#### Delayed-Feedback-Problem\n* [2008 (KDD) Learning Classifiers from Only Positive and Unlabeled Data](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2008%20%28KDD%29%20Learning%20Classifiers%20from%20Only%20Positive%20and%20Unlabeled%20Data.pdf) \u003Cbr \u002F>\n* [2014 (Criteo) (KDD) [DFM] Modeling Delayed Feedback in Display Advertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2014%20%28Criteo%29%20%28KDD%29%20%5BDFM%5D%20Modeling%20Delayed%20Feedback%20in%20Display%20Advertising.pdf) \u003Cbr \u002F>\n* [2018 (Arxiv) [NoDeF] A Nonparametric Delayed Feedback Model for Conversion Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2018%20%28Arxiv%29%20%5BNoDeF%5D%20A%20Nonparametric%20Delayed%20Feedback%20Model%20for%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (Twitter) (RecSys) Addressing Delayed Feedback for Continuous Training with Neural Networks in CTR prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2019%20%28Twitter%29%20%28RecSys%29%20Addressing%20Delayed%20Feedback%20for%20Continuous%20Training%20with%20Neural%20Networks%20in%20CTR%20prediction.pdf) \u003Cbr \u002F>\n* [2020 (AdKDD) Delayed Feedback Model with Negative Binomial Regression for Multiple Conversions](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2020%20%28AdKDD%29%20Delayed%20Feedback%20Model%20with%20Negative%20Binomial%20Regression%20for%20Multiple%20Conversions.pdf) \u003Cbr \u002F>\n* [2020 (JD) (IJCAI) [TS-DL] An Attention-based Model for Conversion Rate Prediction with Delayed Feedback via Post-click Calibration](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2020%20%28JD%29%20%28IJCAI%29%20%5BTS-DL%5D%20An%20Attention-based%20Model%20for%20Conversion%20Rate%20Prediction%20with%20Delayed%20Feedback%20via%20Post-click%20Calibration.pdf) \u003Cbr \u002F>\n* [2020 (SIGIR) [DLA-DF] Dual Learning Algorithm for Delayed Conversions](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2020%20%28SIGIR%29%20%5BDLA-DF%5D%20Dual%20Learning%20Algorithm%20for%20Delayed%20Conversions.pdf) \u003Cbr \u002F>\n* [2020 (WWW) [FSIW] A Feedback Shift Correction in Predicting Conversion Rates under Delayed Feedback](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2020%20%28WWW%29%20%5BFSIW%5D%20A%20Feedback%20Shift%20Correction%20in%20Predicting%20Conversion%20Rates%20under%20Delayed%20Feedback.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (AAAI) [ES-DFM] Capturing Delayed Feedback in Conversion Rate Prediction via Elapsed-Time Sampling](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Alibaba%29%20%28AAAI%29%20%5BES-DFM%5D%20Capturing%20Delayed%20Feedback%20in%20Conversion%20Rate%20Prediction%20via%20Elapsed-Time%20Sampling.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (AAAI) [ESDF] Delayed Feedback Modeling for the Entire Space Conversion Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Alibaba%29%20%28AAAI%29%20%5BESDF%5D%20Delayed%20Feedback%20Modeling%20for%20the%20Entire%20Space%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (Arxiv) [Defer] Real Negatives Matter - Continuous Training with Real Negatives for Delayed Feedback Modeling](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Alibaba%29%20%28Arxiv%29%20%5BDefer%5D%20Real%20Negatives%20Matter%20-%20Continuous%20Training%20with%20Real%20Negatives%20for%20Delayed%20Feedback%20Modeling.pdf) \u003Cbr \u002F>\n* [2021 (Google) (Arxiv) Handling many conversions per click in modeling delayed feedback](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Google%29%20%28Arxiv%29%20Handling%20many%20conversions%20per%20click%20in%20modeling%20delayed%20feedback.pdf) \u003Cbr \u002F>\n* [2021 (Tencent) (SIGIR) Counterfactual Reward Modification for Streaming Recommendation with Delayed Feedback](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Tencent%29%20%28SIGIR%29%20Counterfactual%20Reward%20Modification%20for%20Streaming%20Recommendation%20with%20Delayed%20Feedback.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (WWW) [DEFUSE] Asymptotically Unbiased Estimation for Delayed Feedback Modeling via Label Correction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2022%20%28Alibaba%29%20%28WWW%29%20%5BDEFUSE%5D%20Asymptotically%20Unbiased%20Estimation%20for%20Delayed%20Feedback%20Modeling%20via%20Label%20Correction.pdf) \u003Cbr \u002F>\n\n#### Distill\n* [2020 (Alibaba) (KDD) *[Privileged Features Distillation] Privileged Features Distillation at Taobao Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDistill\u002F2020%20%28Alibaba%29%20%28KDD%29%20%2A%5BPrivileged%20Features%20Distillation%5D%20Privileged%20Features%20Distillation%20at%20Taobao%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024 (Google) Self-Auxiliary Distillation for Sample Efficient Learning in Google-Scale Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDistill\u002F2024%20%28Google%29%20Self-Auxiliary%20Distillation%20for%20Sample%20Efficient%20Learning%20in%20Google-Scale%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) (KDD) [HA-PFD] Hardness-aware Privileged Features Distillation with Latent Alignment for CVR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDistill\u002F2025%20%28Bytedance%29%20%28KDD%29%20%5BHA-PFD%5D%20Hardness-aware%20Privileged%20Features%20Distillation%20with%20Latent%20Alignment%20for%20CVR%20Prediction.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) [MIKD] Mutual Information-aware Knowledge Distillation for Short Video Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDistill\u002F2025%20%28Kuaishou%29%20%5BMIKD%5D%20Mutual%20Information-aware%20Knowledge%20Distillation%20for%20Short%20Video%20Recommendation.pdf) \u003Cbr \u002F>\n\n#### Experiment\n* [2010 (Google) Overlapping Experiment Infrastructure - More, Better, Faster Experimentation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FExperiment\u002F2010%20%28Google%29%20Overlapping%20Experiment%20Infrastructure%20-%20More%2C%20Better%2C%20Faster%20Experimentation.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) OptimizedCost perClickin TaobaoDisplayAdvertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FExperiment\u002F2019%20%28Alibaba%29%20%28KDD%29%20OptimizedCost%20perClickin%20TaobaoDisplayAdvertising.pdf) \u003Cbr \u002F>\n* [2022 (Google) (KDD) Scale Calibration of Deep Ranking Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FExperiment\u002F2022%20%28Google%29%20%28KDD%29%20Scale%20Calibration%20of%20Deep%20Ranking%20Models.pdf) \u003Cbr \u002F>\n\n#### Feature-Crossing\n* [2010 (ICDM) [FM] Factorization machines](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2010%20%28ICDM%29%20%5BFM%5D%20Factorization%20machines.pdf) \u003Cbr \u002F>\n* [2013 (Google) (KDD) [LR] Ad Click Prediction - a View from the Trenches](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2013%20%28Google%29%20%28KDD%29%20%5BLR%5D%20Ad%20Click%20Prediction%20-%20a%20View%20from%20the%20Trenches.pdf) \u003Cbr \u002F>\n* [2016 (Arxiv) [PNN] Product-based Neural Networks for User Response Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28Arxiv%29%20%5BPNN%5D%20Product-based%20Neural%20Networks%20for%20User%20Response%20Prediction.pdf) \u003Cbr \u002F>\n* [2016 (Criteo) (Recsys) [FFM] Field-aware Factorization Machines for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28Criteo%29%20%28Recsys%29%20%5BFFM%5D%20Field-aware%20Factorization%20Machines%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2016 (ECIR) [FNN] Deep Learning over Multi-field Categorical Data – A Case Study on User Response Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28ECIR%29%20%5BFNN%5D%20Deep%20Learning%20over%20Multi-field%20Categorical%20Data%20%E2%80%93%20A%20Case%20Study%20on%20User%20Response%20Prediction.pdf) \u003Cbr \u002F>\n* [2016 (KDD) [Deepintent] Deepintent - Learning attentions for online advertising with recurrent neural networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28KDD%29%20%5BDeepintent%5D%20Deepintent%20-%20Learning%20attentions%20for%20online%20advertising%20with%20recurrent%20neural%20networks.pdf) \u003Cbr \u002F>\n* [2016 (Microsoft) (KDD) [Deep Crossing] Deep Crossing - Web-scale modeling without manually crafted combinatorial features](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28Microsoft%29%20%28KDD%29%20%5BDeep%20Crossing%5D%20Deep%20Crossing%20-%20Web-scale%20modeling%20without%20manually%20crafted%20combinatorial%20features.pdf) \u003Cbr \u002F>\n* [2017 (Google) (ADKDD) [DCN] Deep & CrossNetwork for Ad Click Predictions](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28Google%29%20%28ADKDD%29%20%5BDCN%5D%20Deep%20%26%20CrossNetwork%20for%20Ad%20Click%20Predictions.pdf) \u003Cbr \u002F>\n* [2017 (Huawei)  (IJCAI) [DeepFM] DeepFM - A Factorization-Machine based Neural Network for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28Huawei%29%20%20%28IJCAI%29%20%5BDeepFM%5D%20DeepFM%20-%20A%20Factorization-Machine%20based%20Neural%20Network%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2017 (IJCAI) [AFM] Attentional Factorization Machines Learning the Weight of Feature Interactions via Attention Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28IJCAI%29%20%5BAFM%5D%20Attentional%20Factorization%20Machines%20Learning%20the%20Weight%20of%20Feature%20Interactions%20via%20Attention%20Networks.pdf) \u003Cbr \u002F>\n* [2017 (SIGIR) [NFM] Neural Factorization Machines for Sparse Predictive Analytics](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28SIGIR%29%20%5BNFM%5D%20Neural%20Factorization%20Machines%20for%20Sparse%20Predictive%20Analytics.pdf) \u003Cbr \u002F>\n* [2017 (WWW) [NCF] Neural Collaborative Filtering](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28WWW%29%20%5BNCF%5D%20Neural%20Collaborative%20Filtering.pdf) \u003Cbr \u002F>\n* [2018 (Google) (WSDM) [Latent Cross] Latent Cross Making Use of Context in Recurrent Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2018%20%28Google%29%20%28WSDM%29%20%5BLatent%20Cross%5D%20Latent%20Cross%20Making%20Use%20of%20Context%20in%20Recurrent%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2018 (KDD) [xDeepFM] xDeepFM - Combining Explicit and Implicit Feature Interactions for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2018%20%28KDD%29%20%5BxDeepFM%5D%20xDeepFM%20-%20Combining%20Explicit%20and%20Implicit%20Feature%20Interactions%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2018 (TOIS) [PNN] Product-Based Neural Networks for User Response Prediction over Multi-Field Categorical Data](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2018%20%28TOIS%29%20%5BPNN%5D%20Product-Based%20Neural%20Networks%20for%20User%20Response%20Prediction%20over%20Multi-Field%20Categorical%20Data.pdf) \u003Cbr \u002F>\n* [2019 (CIKM) ** [AutoInt] AutoInt - Automatic Feature Interaction Learning via Self-Attentive Neural Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2019%20%28CIKM%29%20%2A%2A%20%5BAutoInt%5D%20AutoInt%20-%20Automatic%20Feature%20Interaction%20Learning%20via%20Self-Attentive%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2019 (Huawei) (WWW) [FGCNN] Feature Generation by Convolutional Neural Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2019%20%28Huawei%29%20%28WWW%29%20%5BFGCNN%5D%20Feature%20Generation%20by%20Convolutional%20Neural%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (Sina) (Arxiv) [FAT-DeepFFM] FAT-DeepFFM - Field Attentive Deep Field-aware Factorization Machine](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2019%20%28Sina%29%20%28Arxiv%29%20%5BFAT-DeepFFM%5D%20FAT-DeepFFM%20-%20Field%20Attentive%20Deep%20Field-aware%20Factorization%20Machine.pdf) \u003Cbr \u002F>\n* [2019 (Tencent) (AAAI) [IFM] Interaction-aware Factorization Machines for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2019%20%28Tencent%29%20%28AAAI%29%20%5BIFM%5D%20Interaction-aware%20Factorization%20Machines%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2020 (Baidu) (KDD) [CAN] Combo-Attention Network for Baidu Video Advertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2020%20%28Baidu%29%20%28KDD%29%20%5BCAN%5D%20Combo-Attention%20Network%20for%20Baidu%20Video%20Advertising.pdf) \u003Cbr \u002F>\n* [2021 (Google) (NIPS) [MLP-Mixer] MLP-Mixer - An all-MLP Architecture for Vision](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2021%20%28Google%29%20%28NIPS%29%20%5BMLP-Mixer%5D%20MLP-Mixer%20-%20An%20all-MLP%20Architecture%20for%20Vision.pdf) \u003Cbr \u002F>\n* [2021 (Google) (WWW) * [DCN V2] DCN V2 - Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2021%20%28Google%29%20%28WWW%29%20%2A%20%5BDCN%20V2%5D%20DCN%20V2%20-%20Improved%20Deep%20%26%20Cross%20Network%20and%20Practical%20Lessons%20for%20Web-scale%20Learning%20to%20Rank%20Systems.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (WSDM) *  [CAN] CAN - Feature Co-Action Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2022%20%28Alibaba%29%20%28WSDM%29%20%2A%20%20%5BCAN%5D%20CAN%20-%20Feature%20Co-Action%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2022 (Meta) ** (Arxiv) DHEN - A Deep and Hierarchical Ensemble Network for Large-Scale Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2022%20%28Meta%29%20%2A%2A%20%28Arxiv%29%20DHEN%20-%20A%20Deep%20and%20Hierarchical%20Ensemble%20Network%20for%20Large-Scale%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (CIKM) * [GDCN] Towards Deeper, Lighter and Interpretable Cross Network for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2023%20%28CIKM%29%20%2A%20%5BGDCN%5D%20Towards%20Deeper%2C%20Lighter%20and%20Interpretable%20Cross%20Network%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Google) ** (Arxiv) [Hiformer] Hiformer - Heterogeneous Feature Interactions Learning with Transformers for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2023%20%28Google%29%20%2A%2A%20%28Arxiv%29%20%5BHiformer%5D%20Hiformer%20-%20Heterogeneous%20Feature%20Interactions%20Learning%20with%20Transformers%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2023 (Sina) (CIKM) [MemoNet] MemoNet - Memorizing All Cross Features’ Representations Efficiently via Multi-Hash Codebook Network for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2023%20%28Sina%29%20%28CIKM%29%20%5BMemoNet%5D%20MemoNet%20-%20Memorizing%20All%20Cross%20Features%E2%80%99%20Representations%20Efficiently%20via%20Multi-Hash%20Codebook%20Network%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2024 (Meta) ** (PMLR) [Wukong] Wukong - Towards a Scaling Law for Large-Scale Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2024%20%28Meta%29%20%2A%2A%20%28PMLR%29%20%5BWukong%5D%20Wukong%20-%20Towards%20a%20Scaling%20Law%20for%20Large-Scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 （LinkedIn) (KDD) [RDCN] LiRank - Industrial Large Scale Ranking Models at LinkedIn](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2024%20%EF%BC%88LinkedIn%29%20%28KDD%29%20%5BRDCN%5D%20LiRank%20-%20Industrial%20Large%20Scale%20Ranking%20Models%20at%20LinkedIn.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) [HHFT] HHFT - Hierarchical Heterogeneous Feature Transformer for Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Alibaba%29%20%5BHHFT%5D%20HHFT%20-%20Hierarchical%20Heterogeneous%20Feature%20Transformer%20for%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (Arxiv) [Pyramid Mixer] Pyramid Mixer - Multi-dimensional Multi-period Interest Modeling for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BPyramid%20Mixer%5D%20Pyramid%20Mixer%20-%20Multi-dimensional%20Multi-period%20Interest%20Modeling%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (Arxiv) [RankMixer] RankMixer - Scaling Up Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (CIKM) [RankMixer] RankMixer - Scaling Up Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Bytedance%29%20%2A%2A%20%28CIKM%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Meta) (CIKM) [InterFormer] InterFormer - Effective Heterogeneous Interaction Learning for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Meta%29%20%28CIKM%29%20%5BInterFormer%5D%20InterFormer%20-%20Effective%20Heterogeneous%20Interaction%20Learning%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2025 (Tencent) (Arxiv) [D-MoE] Enhancing CTR Prediction with De-correlated Expert Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Tencent%29%20%28Arxiv%29%20%5BD-MoE%5D%20Enhancing%20CTR%20Prediction%20with%20De-correlated%20Expert%20Networks.pdf) \u003Cbr \u002F>\n* [2025 （Alibaba) [FAT] From Scaling to Structured Expressivity - Rethinking Transformers for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%EF%BC%88Alibaba%29%20%5BFAT%5D%20From%20Scaling%20to%20Structured%20Expressivity%20-%20Rethinking%20Transformers%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2025 （Bytedance) (Arxiv) [OneTrans] OneTrans - Unified Feature Interaction and Sequence Modeling with One Transformer in Industrial Recommender](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%EF%BC%88Bytedance%29%20%28Arxiv%29%20%5BOneTrans%5D%20OneTrans%20-%20Unified%20Feature%20Interaction%20and%20Sequence%20Modeling%20with%20One%20Transformer%20in%20Industrial%20Recommender.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [MixFormer] MixFormer - Co-Scaling Up Dense and Sequence in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BMixFormer%5D%20MixFormer%20-%20Co-Scaling%20Up%20Dense%20and%20Sequence%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [TokenMixer-Large] TokenMixer-Large - Scaling Up Large Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BTokenMixer-Large%5D%20TokenMixer-Large%20-%20Scaling%20Up%20Large%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [Zenith] Zenith - Scaling up Ranking Models for Billion-scale Livestreaming Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BZenith%5D%20Zenith%20-%20Scaling%20up%20Ranking%20Models%20for%20Billion-scale%20Livestreaming%20Recommendation.pdf) \u003Cbr \u002F>\n* [2026 (Kuaishou) (Arxiv) [UniMixer] UniMixer - A Unified Architecture for Scaling Laws in Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2026%20%28Kuaishou%29%20%28Arxiv%29%20%5BUniMixer%5D%20UniMixer%20-%20A%20Unified%20Architecture%20for%20Scaling%20Laws%20in%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n\n#### Feature_Importance\n* [2022 (Kuaishou) (Arxiv) [LPFS] LPFS - Learnable Polarizing Feature Selection for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature_Importance\u002F2022%20%28Kuaishou%29%20%28Arxiv%29%20%5BLPFS%5D%20LPFS%20-%20Learnable%20Polarizing%20Feature%20Selection%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2024 (Huawei)(KDD) ERASE - Benchmarking Feature Selection Methods for Deep Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature_Importance\u002F2024%20%28Huawei%29%28KDD%29%20ERASE%20-%20Benchmarking%20Feature%20Selection%20Methods%20for%20Deep%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n\n#### Gating\n* [2014 (TASLP) * [LHUC] Learning Hidden Unit Contributions for Unsupervised Acoustic Model Adaptation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2014%20%28TASLP%29%20%2A%20%5BLHUC%5D%20Learning%20Hidden%20Unit%20Contributions%20for%20Unsupervised%20Acoustic%20Model%20Adaptation.pdf) \u003Cbr \u002F>\n* [2018 (CVPR) * [SENet] Squeeze-and-Excitation Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2018%20%28CVPR%29%20%2A%20%5BSENet%5D%20Squeeze-and-Excitation%20Networks.pdf) \u003Cbr \u002F>\n* [2019 (Sina) (Recsys) [FiBiNET] FiBiNET - combining feature importance and bilinear feature interaction for click-through rate prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2019%20%28Sina%29%20%28Recsys%29%20%5BFiBiNET%5D%20FiBiNET%20-%20combining%20feature%20importance%20and%20bilinear%20feature%20interaction%20for%20click-through%20rate%20prediction.pdf) \u003Cbr \u002F>\n* [2020 (Sina) (Arxiv) [GateNet] GateNet - Gating-Enhanced Deep Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2020%20%28Sina%29%20%28Arxiv%29%20%5BGateNet%5D%20GateNet%20-%20Gating-Enhanced%20Deep%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Sina) (Arxiv) [ContextNet] ContextNet - A Click-Through Rate Prediction Framework Using Contextual information to Refine Feature Embedding](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2021%20%28Sina%29%20%28Arxiv%29%20%5BContextNet%5D%20ContextNet%20-%20A%20Click-Through%20Rate%20Prediction%20Framework%20Using%20Contextual%20information%20to%20Refine%20Feature%20Embedding.pdf) \u003Cbr \u002F>\n* [2021 (Sina) （DLP-KDD) [MaskNet] MaskNet - Introducing Feature-Wise Multiplication to CTR Ranking Models by Instance-Guided Mask](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2021%20%28Sina%29%20%EF%BC%88DLP-KDD%29%20%5BMaskNet%5D%20MaskNet%20-%20Introducing%20Feature-Wise%20Multiplication%20to%20CTR%20Ranking%20Models%20by%20Instance-Guided%20Mask.pdf) \u003Cbr \u002F>\n* [2023 (Kuaishou) (KDD) [PEPNet] PEPNet - Parameter and Embedding Personalized Network for Infusing with Personalized Prior Information](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2023%20%28Kuaishou%29%20%28KDD%29%20%5BPEPNet%5D%20PEPNet%20-%20Parameter%20and%20Embedding%20Personalized%20Network%20for%20Infusing%20with%20Personalized%20Prior%20Information.pdf) \u003Cbr \u002F>\n* [2023 (Sina) (CIKM) [FiBiNet++] FiBiNet++ - Reducing Model Size by Low Rank Feature Interaction Layer for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2023%20%28Sina%29%20%28CIKM%29%20%5BFiBiNet%2B%2B%5D%20FiBiNet%2B%2B%20-%20Reducing%20Model%20Size%20by%20Low%20Rank%20Feature%20Interaction%20Layer%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) [ADS] Adaptive Domain Scaling for Personalized Sequential Modeling in Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2025%20%28Bytedance%29%20%5BADS%5D%20Adaptive%20Domain%20Scaling%20for%20Personalized%20Sequential%20Modeling%20in%20Recommenders.pdf) \u003Cbr \u002F>\n\n#### LLM_Ranking\n* [2019 (CIKM)  [AutoInt] AutoInt -Automatic Feature Interaction Learning via Self-Attentive Neural Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2019%20%28CIKM%29%20%20%5BAutoInt%5D%20AutoInt%20-Automatic%20Feature%20Interaction%20Learning%20via%20Self-Attentive%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2020 (Arxiv) Scaling Laws for Neural Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2020%20%28Arxiv%29%20Scaling%20Laws%20for%20Neural%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2021 (Baidu) (KDD) Pre-trained Language Model based Ranking in Baidu Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2021%20%28Baidu%29%20%28KDD%29%20Pre-trained%20Language%20Model%20based%20Ranking%20in%20Baidu%20Search.pdf) \u003Cbr \u002F>\n* [2021 (Google) (Arxiv) [MLP-Mixer] MLP-Mixer - An all-MLP Architecture for Vision](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2021%20%28Google%29%20%28Arxiv%29%20%5BMLP-Mixer%5D%20MLP-Mixer%20-%20An%20all-MLP%20Architecture%20for%20Vision.pdf) \u003Cbr \u002F>\n* [2022 (Meta) ** (Arxiv) DHEN - A Deep and Hierarchical Ensemble Network for Large-Scale Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2022%20%28Meta%29%20%2A%2A%20%28Arxiv%29%20DHEN%20-%20A%20Deep%20and%20Hierarchical%20Ensemble%20Network%20for%20Large-Scale%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Arxiv) [E4SRec] E4SRec - An Elegant Effective Efficient Extensible Solution of Large Language Models for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2023%20%28Arxiv%29%20%5BE4SRec%5D%20E4SRec%20-%20An%20Elegant%20Effective%20Efficient%20Extensible%20Solution%20of%20Large%20Language%20Models%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023 (Google) ** (Arxiv) [Hiformer] Hiformer - Heterogeneous Feature Interactions Learning with Transformers for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2023%20%28Google%29%20%2A%2A%20%28Arxiv%29%20%5BHiformer%5D%20Hiformer%20-%20Heterogeneous%20Feature%20Interactions%20Learning%20with%20Transformers%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2024 (Alibaba) (Arxiv) [BAHE] Breaking the Length Barrier - LLM-Enhanced CTR Prediction in Long Textual User Behaviors](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Alibaba%29%20%28Arxiv%29%20%5BBAHE%5D%20Breaking%20the%20Length%20Barrier%20-%20LLM-Enhanced%20CTR%20Prediction%20in%20Long%20Textual%20User%20Behaviors.pdf) \u003Cbr \u002F>\n* [2024 (Bytedance) (Arxiv) [HLLM] HLLM - Enhancing Sequential Recommendations via Hierarchical Large Language Models for Item and User Modeling](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Bytedance%29%20%28Arxiv%29%20%5BHLLM%5D%20HLLM%20-%20Enhancing%20Sequential%20Recommendations%20via%20Hierarchical%20Large%20Language%20Models%20for%20Item%20and%20User%20Modeling.pdf) \u003Cbr \u002F>\n* [2024 (Google) (Arxiv) LLMs for User Interest Exploration in Large-scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Google%29%20%28Arxiv%29%20LLMs%20for%20User%20Interest%20Exploration%20in%20Large-scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2024 (Google) (Arxiv) [CALRec] CALRec - Contrastive Alignment of Generative LLMs for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Google%29%20%28Arxiv%29%20%5BCALRec%5D%20CALRec%20-%20Contrastive%20Alignment%20of%20Generative%20LLMs%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 (Google) (ICLR) From Sparse to Soft Mixtures of Experts](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Google%29%20%28ICLR%29%20From%20Sparse%20to%20Soft%20Mixtures%20of%20Experts.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (Arxiv) [LEARN] LEARN - Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Application](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Kuaishou%29%20%28Arxiv%29%20%5BLEARN%5D%20LEARN%20-%20Knowledge%20Adaptation%20from%20Large%20Language%20Model%20to%20Recommendation%20for%20Practical%20Industrial%20Application.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (KDD) [NAR4Rec] Non-autoregressive Generative Models for Reranking Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Kuaishou%29%20%28KDD%29%20%5BNAR4Rec%5D%20Non-autoregressive%20Generative%20Models%20for%20Reranking%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 (Meituan) (Arxiv) [SRP4CTR] Enhancing CTR Prediction through Sequential Recommendation Pre-training - Introducing the SRP4CTR Framework](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meituan%29%20%28Arxiv%29%20%5BSRP4CTR%5D%20Enhancing%20CTR%20Prediction%20through%20Sequential%20Recommendation%20Pre-training%20-%20Introducing%20the%20SRP4CTR%20Framework.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) ** [GR] Actions Speak Louder than Words - Trillion-Parameter Sequential Transducers for Generative Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meta%29%20%28Arxiv%29%20%2A%2A%20%5BGR%5D%20Actions%20Speak%20Louder%20than%20Words%20-%20Trillion-Parameter%20Sequential%20Transducers%20for%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) Unifying Generative and Dense Retrieval for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meta%29%20%28Arxiv%29%20Unifying%20Generative%20and%20Dense%20Retrieval%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) [SUM] Scaling User Modeling - Large-scale Online User Representations for Ads Personalization in Meta](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meta%29%20%28Arxiv%29%20%5BSUM%5D%20Scaling%20User%20Modeling%20-%20Large-scale%20Online%20User%20Representations%20for%20Ads%20Personalization%20in%20Meta.pdf) \u003Cbr \u002F>\n* [2024 (Meta) ** (PMLR) [Wukong] Wukong - Towards a Scaling Law for Large-Scale Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meta%29%20%2A%2A%20%28PMLR%29%20%5BWukong%5D%20Wukong%20-%20Towards%20a%20Scaling%20Law%20for%20Large-Scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025  (Kuaishou) (Arxiv)[OneRec] OneRec - Unifying Retrieve and Rank with Generative Recommender and Preference Alignment](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%20%28Kuaishou%29%20%28Arxiv%29%5BOneRec%5D%20OneRec%20-%20Unifying%20Retrieve%20and%20Rank%20with%20Generative%20Recommender%20and%20Preference%20Alignment.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) Unlocking Scaling Law in Industrial Recommendation Systems with a Three-step Paradigm based Large User Model](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20Unlocking%20Scaling%20Law%20in%20Industrial%20Recommendation%20Systems%20with%20a%20Three-step%20Paradigm%20based%20Large%20User%20Model.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [HeterRec] Hierarchical Causal Transformer with Heterogeneous Information for Expandable Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BHeterRec%5D%20Hierarchical%20Causal%20Transformer%20with%20Heterogeneous%20Information%20for%20Expandable%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [LREA] Efficient Long Sequential Low-rank Adaptive Attention for Click-through rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BLREA%5D%20Efficient%20Long%20Sequential%20Low-rank%20Adaptive%20Attention%20for%20Click-through%20rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [URM] Large Language Models Are Universal Recommendation Learners](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BURM%5D%20Large%20Language%20Models%20Are%20Universal%20Recommendation%20Learners.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (KDD) [GPSD] Scaling Transformers for Discriminative Recommendation via Generative Pretraining](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28KDD%29%20%5BGPSD%5D%20Scaling%20Transformers%20for%20Discriminative%20Recommendation%20via%20Generative%20Pretraining.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (WWW) Explainable LLM-driven Multi-dimensional Distillation for E-Commerce Relevance Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28WWW%29%20Explainable%20LLM-driven%20Multi-dimensional%20Distillation%20for%20E-Commerce%20Relevance%20Learning.pdf) \u003Cbr \u002F>\n* [2025 (Amazon) (Arxiv) SynerGen - Contextualized Generative Recommender for Unified Search and Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Amazon%29%20%28Arxiv%29%20SynerGen%20-%20Contextualized%20Generative%20Recommender%20for%20Unified%20Search%20and%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Arxiv) (Pinterest) [PinRec] PinRec - Outcome-Conditioned, Multi-Token Generative Retrieval for Industry-Scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Arxiv%29%20%28Pinterest%29%20%5BPinRec%5D%20PinRec%20-%20Outcome-Conditioned%2C%20Multi-Token%20Generative%20Retrieval%20for%20Industry-Scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Arxiv) (Xiaohongshu) [GenRank] Towards Large-scale Generative Ranking](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Arxiv%29%20%28Xiaohongshu%29%20%5BGenRank%5D%20Towards%20Large-scale%20Generative%20Ranking.pdf) \u003Cbr \u002F>\n* [2025 (Baidu) (Arxiv) [COBRA] Sparse Meets Dense -Unified Generative Recommendations with Cascaded Sparse-Dense Representations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Baidu%29%20%28Arxiv%29%20%5BCOBRA%5D%20Sparse%20Meets%20Dense%20-Unified%20Generative%20Recommendations%20with%20Cascaded%20Sparse-Dense%20Representations.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) (Arxiv) [RankMixer] RankMixer - Scaling Up Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Bytedance%29%20%28Arxiv%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (Arxiv) [LONGER] LONGER - Scaling Up Long Sequence Modeling in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BLONGER%5D%20LONGER%20-%20Scaling%20Up%20Long%20Sequence%20Modeling%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (CIKM) [RankMixer] RankMixer - Scaling Up Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Bytedance%29%20%2A%2A%20%28CIKM%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Google) (Arxiv) User Feedback Alignment for LLM-powered Exploration in Large-scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Google%29%20%28Arxiv%29%20User%20Feedback%20Alignment%20for%20LLM-powered%20Exploration%20in%20Large-scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Google) (Arxiv) [STAR] STAR - A Simple Training-free Approach for Recommendations using Large Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Google%29%20%28Arxiv%29%20%5BSTAR%5D%20STAR%20-%20A%20Simple%20Training-free%20Approach%20for%20Recommendations%20using%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2025 (Google) ** (Arxiv) [Hiformer] Hiformer - Heterogeneous Feature Interactions Learning with Transformers for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Google%29%20%2A%2A%20%28Arxiv%29%20%5BHiformer%5D%20Hiformer%20-%20Heterogeneous%20Feature%20Interactions%20Learning%20with%20Transformers%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [GenSAR] Unified Generative Search and Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BGenSAR%5D%20Unified%20Generative%20Search%20and%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [LARM] LLM-Alignment Live-Streaming Recommendationpdf](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BLARM%5D%20LLM-Alignment%20Live-Streaming%20Recommendationpdf.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [LEARN] LEARN - Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Application](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BLEARN%5D%20LEARN%20-%20Knowledge%20Adaptation%20from%20Large%20Language%20Model%20to%20Recommendation%20for%20Practical%20Industrial%20Application.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneLoc] OneLoc - Geo-Aware Generative Recommender Systems for Local Life Service](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneLoc%5D%20OneLoc%20-%20Geo-Aware%20Generative%20Recommender%20Systems%20for%20Local%20Life%20Service.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneRec-V2] OneRec Technical Report v2](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneRec-V2%5D%20OneRec%20Technical%20Report%20v2.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneRec] OneRec - Unifying Retrieve and Rank with Generative Recommender and Preference Alignment](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneRec%5D%20OneRec%20-%20Unifying%20Retrieve%20and%20Rank%20with%20Generative%20Recommender%20and%20Preference%20Alignment.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneRec] OneRec Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneRec%5D%20OneRec%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneSearch] OneSearch - A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneSearch%5D%20OneSearch%20-%20A%20Preliminary%20Exploration%20of%20the%20Unified%20End-to-End%20Generative%20Framework%20for%20E-commerce%20Search.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneSug] OneSug - The Unified End-to-End Generative Framework for E-commerce Query Suggestion](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneSug%5D%20OneSug%20-%20The%20Unified%20End-to-End%20Generative%20Framework%20for%20E-commerce%20Query%20Suggestion.pdf) \u003Cbr \u002F>\n* [2025 (Meituan) (Arxiv) [DFGR] Action is All You Need - Dual-Flow Generative Ranking Network for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Meituan%29%20%28Arxiv%29%20%5BDFGR%5D%20Action%20is%20All%20You%20Need%20-%20Dual-Flow%20Generative%20Ranking%20Network%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Meituan) (Arxiv) [MTGR] MTGR - Industrial-Scale Generative Recommendation Framework in Meituan](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Meituan%29%20%28Arxiv%29%20%5BMTGR%5D%20MTGR%20-%20Industrial-Scale%20Generative%20Recommendation%20Framework%20in%20Meituan.pdf) \u003Cbr \u002F>\n* [2025 (Meituan) (Arxiv) [UniROM] One Model to Rank Them All - Unifying Online Advertising with End-to-End Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Meituan%29%20%28Arxiv%29%20%5BUniROM%5D%20One%20Model%20to%20Rank%20Them%20All%20-%20Unifying%20Online%20Advertising%20with%20End-to-End%20Learning.pdf) \u003Cbr \u002F>\n* [2025 (Meta) (Arxiv) [HyperCast] Realizing Scaling Laws in Recommender Systems - A Foundation–Expert Paradigm for Hyperscale Model Deployment](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Meta%29%20%28Arxiv%29%20%5BHyperCast%5D%20Realizing%20Scaling%20Laws%20in%20Recommender%20Systems%20-%20A%20Foundation%E2%80%93Expert%20Paradigm%20for%20Hyperscale%20Model%20Deployment.pdf) \u003Cbr \u002F>\n* [2025 (Microsoft) (KDD)Towards Web-scale Recommendations with LLMs - From Quality-aware Ranking to Candidate Generation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Microsoft%29%20%28KDD%29Towards%20Web-scale%20Recommendations%20with%20LLMs%20-%20From%20Quality-aware%20Ranking%20to%20Candidate%20Generation.pdf) \u003Cbr \u002F>\n* [2025 (Pinterest) (Arxiv) [PinFM] PinFM - Foundation Model for User Activity Sequences at a Billion-scale Visual Discovery Platform](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Pinterest%29%20%28Arxiv%29%20%5BPinFM%5D%20PinFM%20-%20Foundation%20Model%20for%20User%20Activity%20Sequences%20at%20a%20Billion-scale%20Visual%20Discovery%20Platform.pdf) \u003Cbr \u002F>\n* [2025 (Shopee) （Arxiv) OnePiece - Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Shopee%29%20%EF%BC%88Arxiv%29%20OnePiece%20-%20Bringing%20Context%20Engineering%20and%20Reasoning%20to%20Industrial%20Cascade%20Ranking%20System.pdf) \u003Cbr \u002F>\n* [2025 (Tencent) (Arxiv) [GPR] GPR - Towards a Generative Pre-trained One-Model Paradigm for Large-Scale Advertising Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Tencent%29%20%28Arxiv%29%20%5BGPR%5D%20GPR%20-%20Towards%20a%20Generative%20Pre-trained%20One-Model%20Paradigm%20for%20Large-Scale%20Advertising%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (eBay)  (Arxiv) LLMDistill4Ads - Using Cross-Encoders to Distill from LLM Signals for Advertiser Keyphrase Recommendations at eBay](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28eBay%29%20%20%28Arxiv%29%20LLMDistill4Ads%20-%20Using%20Cross-Encoders%20to%20Distill%20from%20LLM%20Signals%20for%20Advertiser%20Keyphrase%20Recommendations%20at%20eBay.pdf) \u003Cbr \u002F>\n* [2025 （Bytedance) (Arxiv) [OneTrans] OneTrans - Unified Feature Interaction and Sequence Modeling with One Transformer in Industrial Recommender](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%EF%BC%88Bytedance%29%20%28Arxiv%29%20%5BOneTrans%5D%20OneTrans%20-%20Unified%20Feature%20Interaction%20and%20Sequence%20Modeling%20with%20One%20Transformer%20in%20Industrial%20Recommender.pdf) \u003Cbr \u002F>\n* [2025 （Bytedance) ** (Arxiv) [STCA] Make It Long, Keep It Fast - End-to-End 10k-Sequence Modeling at Billion Scale on Douyin](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%EF%BC%88Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BSTCA%5D%20Make%20It%20Long%2C%20Keep%20It%20Fast%20-%20End-to-End%2010k-Sequence%20Modeling%20at%20Billion%20Scale%20on%20Douyin.pdf) \u003Cbr \u002F>\n* [2026 (Alibaba) (Arxiv) [EST] EST - Towards Efficient Scaling Laws in Click-Through Rate Prediction via Unified Modeling](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Alibaba%29%20%28Arxiv%29%20%5BEST%5D%20EST%20-%20Towards%20Efficient%20Scaling%20Laws%20in%20Click-Through%20Rate%20Prediction%20via%20Unified%20Modeling.pdf) \u003Cbr \u002F>\n* [2026 (Alibaba) (Arxiv) [SORT] SORT - A Systematically Optimized Ranking Transformer for Industrial-scale Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Alibaba%29%20%28Arxiv%29%20%5BSORT%5D%20SORT%20-%20A%20Systematically%20Optimized%20Ranking%20Transformer%20for%20Industrial-scale%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [HyFormer] HyFormer - Revisiting the Roles of Sequence Modeling and Feature Interaction in CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BHyFormer%5D%20HyFormer%20-%20Revisiting%20the%20Roles%20of%20Sequence%20Modeling%20and%20Feature%20Interaction%20in%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [MixFormer] MixFormer - Co-Scaling Up Dense and Sequence in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BMixFormer%5D%20MixFormer%20-%20Co-Scaling%20Up%20Dense%20and%20Sequence%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [TokenMixer-Large] TokenMixer-Large - Scaling Up Large Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BTokenMixer-Large%5D%20TokenMixer-Large%20-%20Scaling%20Up%20Large%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [UG-Sep] Compute Only Once - UG-Separation for Efficient Large Recommendation Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BUG-Sep%5D%20Compute%20Only%20Once%20-%20UG-Separation%20for%20Efficient%20Large%20Recommendation%20Models.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [Zenith] Zenith - Scaling up Ranking Models for Billion-scale Livestreaming Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BZenith%5D%20Zenith%20-%20Scaling%20up%20Ranking%20Models%20for%20Billion-scale%20Livestreaming%20Recommendation.pdf) \u003Cbr \u002F>\n* [2026 (Kuaishou) (Arxiv) [GR4AD] Generative Recommendation for Large-Scale Advertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Kuaishou%29%20%28Arxiv%29%20%5BGR4AD%5D%20Generative%20Recommendation%20for%20Large-Scale%20Advertising.pdf) \u003Cbr \u002F>\n* [2026 (Meta) (Arxiv) [Kunlun] Kunlun - Establishing Scaling Laws for Massive-Scale Recommendation Systems through Unified Architecture Design](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Meta%29%20%28Arxiv%29%20%5BKunlun%5D%20Kunlun%20-%20Establishing%20Scaling%20Laws%20for%20Massive-Scale%20Recommendation%20Systems%20through%20Unified%20Architecture%20Design.pdf) \u003Cbr \u002F>\n* [2026 (Meta) (Arxiv) [LLaTTE] LLaTTE - Scaling Laws for Multi-Stage Sequence Modeling in Large-Scale Ads Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Meta%29%20%28Arxiv%29%20%5BLLaTTE%5D%20LLaTTE%20-%20Scaling%20Laws%20for%20Multi-Stage%20Sequence%20Modeling%20in%20Large-Scale%20Ads%20Recommendation.pdf) \u003Cbr \u002F>\n* [2026 (Meta) (Arxiv) [ULTRA-HSTU] Bending the Scaling Law Curve in Large-Scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Meta%29%20%28Arxiv%29%20%5BULTRA-HSTU%5D%20Bending%20the%20Scaling%20Law%20Curve%20in%20Large-Scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2026 （Bytedance) (Arxiv) [MDL] MDL - A Unified Multi-Distribution Learner in Large-scale Industrial Recommendation through Tokenization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%EF%BC%88Bytedance%29%20%28Arxiv%29%20%5BMDL%5D%20MDL%20-%20A%20Unified%20Multi-Distribution%20Learner%20in%20Large-scale%20Industrial%20Recommendation%20through%20Tokenization.pdf) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002FLLM_MultiModal_Ranking) \u003Cbr \u002F>\n\n#### Loss\n* [2015 (Twitter) (KDD) Click-through Prediction for Advertising in Twitter Timeline](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2015%20%28Twitter%29%20%28KDD%29%20Click-through%20Prediction%20for%20Advertising%20in%20Twitter%20Timeline.pdf) \u003Cbr \u002F>\n* [2022 (Google) (KDD) Scale Calibration of Deep Ranking Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2022%20%28Google%29%20%28KDD%29%20Scale%20Calibration%20of%20Deep%20Ranking%20Models.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (KDD) Joint Optimization of Ranking and Calibration with Contextualized Hybrid Model](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2023%20%28Alibaba%29%20%28KDD%29%20Joint%20Optimization%20of%20Ranking%20and%20Calibration%20with%20Contextualized%20Hybrid%20Model.pdf) \u003Cbr \u002F>\n* [2023 (Google) (CIKM) Regression Compatible Listwise Objectives for Calibrated Ranking with Binary Relevance](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2023%20%28Google%29%20%28CIKM%29%20Regression%20Compatible%20Listwise%20Objectives%20for%20Calibrated%20Ranking%20with%20Binary%20Relevance.pdf) \u003Cbr \u002F>\n* [2024 (Tencent) (KDD) Understanding the Ranking Loss for Recommendation with Sparse User Feedback](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2024%20%28Tencent%29%20%28KDD%29%20Understanding%20the%20Ranking%20Loss%20for%20Recommendation%20with%20Sparse%20User%20Feedback.pdf) \u003Cbr \u002F>\n\n#### Multi-Modal\n* [2018 (Alibaba) (CIKM) [Image CTR] Image Matters - Visually Modeling User Behaviors Using Advanced Model Server](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-Modal\u002F2018%20%28Alibaba%29%20%28CIKM%29%20%5BImage%20CTR%5D%20Image%20Matters%20-%20Visually%20Modeling%20User%20Behaviors%20Using%20Advanced%20Model%20Server.pdf) \u003Cbr \u002F>\n* [2024 (Alibaba) (CIKM) Enhancing Taobao Display Advertising with Multimodal Representations - Challenges, Approaches and Insights](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-Modal\u002F2024%20%28Alibaba%29%20%28CIKM%29%20Enhancing%20Taobao%20Display%20Advertising%20with%20Multimodal%20Representations%20-%20Challenges%2C%20Approaches%20and%20Insights.pdf) \u003Cbr \u002F>\n* [2026 (Alibaba) (WSDM) [MOON] MOON - Generative MLLM-based Multimodal Representation Learning for E-commerce Product Understanding](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-Modal\u002F2026%20%28Alibaba%29%20%28WSDM%29%20%5BMOON%5D%20MOON%20-%20Generative%20MLLM-based%20Multimodal%20Representation%20Learning%20for%20E-commerce%20Product%20Understanding.pdf) \u003Cbr \u002F>\n\n#### Multi-domain-Multi-Scenario\n* [2014 (TASLP) * [LHUC] Learning Hidden Unit Contributions for Unsupervised Acoustic Model Adaptation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2014%20%28TASLP%29%20%2A%20%5BLHUC%5D%20Learning%20Hidden%20Unit%20Contributions%20for%20Unsupervised%20Acoustic%20Model%20Adaptation.pdf) \u003Cbr \u002F>\n* [2015 (Microsoft) (WWW) A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2015%20%28Microsoft%29%20%28WWW%29%20A%20Multi-View%20Deep%20Learning%20Approach%20for%20Cross%20Domain%20User%20Modeling%20in%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2018 (Google) (KDD) ** [MMoE] Modeling task relationships in multi-task learning with multi-gate mixture-of-experts](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2018%20%28Google%29%20%28KDD%29%20%2A%2A%20%5BMMoE%5D%20Modeling%20task%20relationships%20in%20multi-task%20learning%20with%20multi-gate%20mixture-of-experts.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (CIKM) [WE-CAN] Cross-domain Attention Network with Wasserstein Regularizers for E-commerce Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2019%20%28Alibaba%29%20%28CIKM%29%20%5BWE-CAN%5D%20Cross-domain%20Attention%20Network%20with%20Wasserstein%20Regularizers%20for%20E-commerce%20Search.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (Arxiv) [SAML] Scenario-aware and Mutual-based approach for Multi-scenario Recommendation in E-Commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2020%20%28Alibaba%29%20%28Arxiv%29%20%5BSAML%5D%20Scenario-aware%20and%20Mutual-based%20approach%20for%20Multi-scenario%20Recommendation%20in%20E-Commerce.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (CIKM) [HMoE] Improving Multi-Scenario Learning to Rank in E-commerce by Exploiting Task Relationships in the Label Space](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2020%20%28Alibaba%29%20%28CIKM%29%20%5BHMoE%5D%20Improving%20Multi-Scenario%20Learning%20to%20Rank%20in%20E-commerce%20by%20Exploiting%20Task%20Relationships%20in%20the%20Label%20Space.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba)(CIKM) [MiNet] MiNet - Mixed Interest Network for Cross-Domain Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2020%20%28Alibaba%29%28CIKM%29%20%5BMiNet%5D%20MiNet%20-%20Mixed%20Interest%20Network%20for%20Cross-Domain%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020 (Tencent) (Recsys) **  [PLE] Progressive Layered Extraction (PLE) - A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2020%20%28Tencent%29%20%28Recsys%29%20%2A%2A%20%20%5BPLE%5D%20Progressive%20Layered%20Extraction%20%28PLE%29%20-%20A%20Novel%20Multi-Task%20Learning%20%28MTL%29%20Model%20for%20Personalized%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (CIKM) * [ZEUS] Self-Supervised Learning on Users’ Spontaneous Behaviors for Multi-Scenario Ranking in E-commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%2A%20%5BZEUS%5D%20Self-Supervised%20Learning%20on%20Users%E2%80%99%20Spontaneous%20Behaviors%20for%20Multi-Scenario%20Ranking%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (CIKM) ** [STAR] One Model to Serve All - Star Topology Adaptive Recommender for Multi-Domain CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%2A%2A%20%5BSTAR%5D%20One%20Model%20to%20Serve%20All%20-%20Star%20Topology%20Adaptive%20Recommender%20for%20Multi-Domain%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Google) (ICLR) HyperGrid Transformers - Towards A Single Model for Multiple Tasks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2021%20%28Google%29%20%28ICLR%29%20HyperGrid%20Transformers%20-%20Towards%20A%20Single%20Model%20for%20Multiple%20Tasks.pdf) \u003Cbr \u002F>\n* [2021 (Kwai) (Arxiv) [POSO] POSO - Personalized Cold Start Modules for Large-scale Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2021%20%28Kwai%29%20%28Arxiv%29%20%5BPOSO%5D%20POSO%20-%20Personalized%20Cold%20Start%20Modules%20for%20Large-scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (CIKM) AdaSparse - Learning Adaptively Sparse Structures for Multi-Domain Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2022%20%28Alibaba%29%20%28CIKM%29%20AdaSparse%20-%20Learning%20Adaptively%20Sparse%20Structures%20for%20Multi-Domain%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (NIPS) ** [APG] APG - Adaptive Parameter Generation Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2022%20%28Alibaba%29%20%28NIPS%29%20%2A%2A%20%5BAPG%5D%20APG%20-%20Adaptive%20Parameter%20Generation%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (CIKM) [HC2] Hybrid Contrastive Constraints for Multi-Scenario Ad Ranking](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BHC2%5D%20Hybrid%20Contrastive%20Constraints%20for%20Multi-Scenario%20Ad%20Ranking.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (CIKM) [MMN] Masked Multi-Domain Network - Multi-Type and Multi-Scenario Conversion Rate Prediction with a Single Model](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BMMN%5D%20Masked%20Multi-Domain%20Network%20-%20Multi-Type%20and%20Multi-Scenario%20Conversion%20Rate%20Prediction%20with%20a%20Single%20Model.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (CIKM) [Rec4Ad] Rec4Ad - A Free Lunch to Mitigate Sample Selection Bias for Ads CTR Prediction in Taobao](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BRec4Ad%5D%20Rec4Ad%20-%20A%20Free%20Lunch%20to%20Mitigate%20Sample%20Selection%20Bias%20for%20Ads%20CTR%20Prediction%20in%20Taobao.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (SIGIR) [MARIA] Multi-Scenario Ranking with Adaptive Feature Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Alibaba%29%20%28SIGIR%29%20%5BMARIA%5D%20Multi-Scenario%20Ranking%20with%20Adaptive%20Feature%20Learning.pdf) \u003Cbr \u002F>\n* [2023 (CIKM) [HAMUR] HAMUR - Hyper Adapter for Multi-Domain Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28CIKM%29%20%5BHAMUR%5D%20HAMUR%20-%20Hyper%20Adapter%20for%20Multi-Domain%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023 (Huawei) (CIKM) [DFFM] DFFM - Domain Facilitated Feature Modeling for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Huawei%29%20%28CIKM%29%20%5BDFFM%5D%20DFFM%20-%20Domain%20Facilitated%20Feature%20Modeling%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Kuaishou) (KDD) *  [PEPNet] PEPNet - Parameter and Embedding Personalized Network for Infusing with Personalized Prior Information](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Kuaishou%29%20%28KDD%29%20%2A%20%20%5BPEPNet%5D%20PEPNet%20-%20Parameter%20and%20Embedding%20Personalized%20Network%20for%20Infusing%20with%20Personalized%20Prior%20Information.pdf) \u003Cbr \u002F>\n* [2023 (Tencent) (KDD) Scenario-Adaptive Feature Interaction for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Tencent%29%20%28KDD%29%20Scenario-Adaptive%20Feature%20Interaction%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2024 (Alibaba) (CIKM) * [MultiLoRA] MultiLoRA - Multi-Directional Low-Rank Adaptation for Multi-Domain Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28Alibaba%29%20%28CIKM%29%20%2A%20%5BMultiLoRA%5D%20MultiLoRA%20-%20Multi-Directional%20Low-Rank%20Adaptation%20for%20Multi-Domain%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 (Alibaba) (RecSys) * [MLoRA] MLoRA - Multi-Domain Low-Rank Adaptive Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28Alibaba%29%20%28RecSys%29%20%2A%20%5BMLoRA%5D%20MLoRA%20-%20Multi-Domain%20Low-Rank%20Adaptive%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (SIGIR) [M3oE] M3oE - Multi-Domain Multi-Task Mixture-of-Experts Recommendation Framework](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28Kuaishou%29%20%28SIGIR%29%20%5BM3oE%5D%20M3oE%20-%20Multi-Domain%20Multi-Task%20Mixture-of-Experts%20Recommendation%20Framework.pdf) \u003Cbr \u002F>\n* [2024 (Tencent) (KDD) [LCN] Cross-Domain LifeLong Sequential Modeling for Online Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28Tencent%29%20%28KDD%29%20%5BLCN%5D%20Cross-Domain%20LifeLong%20Sequential%20Modeling%20for%20Online%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2024 (WSDM) Exploring Adapter-based Transfer Learning for Recommender Systems - Empirical Studies and Practical Insights](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28WSDM%29%20Exploring%20Adapter-based%20Transfer%20Learning%20for%20Recommender%20Systems%20-%20Empirical%20Studies%20and%20Practical%20Insights.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (KDD) [HoME] HoME - Hierarchy of Multi-Gate Experts for Multi-Task Learning at Kuaishou](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2025%20%28Kuaishou%29%20%28KDD%29%20%5BHoME%5D%20HoME%20-%20Hierarchy%20of%20Multi-Gate%20Experts%20for%20Multi-Task%20Learning%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n\n#### Multi-task\n* [(2018) (ICML) GradNorm - Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F%282018%29%20%28ICML%29%20GradNorm%20-%20Gradient%20Normalization%20for%20Adaptive%20Loss%20Balancing%20in%20Deep%20Multitask%20Networks.pdf) \u003Cbr \u002F>\n* [2014 (TASLP) [LHUC] Learning Hidden Unit Contributions for Unsupervised Acoustic Model Adaptation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2014%20%28TASLP%29%20%5BLHUC%5D%20Learning%20Hidden%20Unit%20Contributions%20for%20Unsupervised%20Acoustic%20Model%20Adaptation.pdf) \u003Cbr \u002F>\n* [2017 (Google) (ICLR) [Sparsely-Gated MOE] Outrageously large neural networks - The sparsely-gated mixture-of-experts layer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2017%20%28Google%29%20%28ICLR%29%20%5BSparsely-Gated%20MOE%5D%20Outrageously%20large%20neural%20networks%20-%20The%20sparsely-gated%20mixture-of-experts%20layer.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (KDD) [DUPN] Perceive Your Users in Depth - Learning Universal User Representations from Multiple E-commerce Tasks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2018%20%28Alibaba%29%20%28KDD%29%20%5BDUPN%5D%20Perceive%20Your%20Users%20in%20Depth%20-%20Learning%20Universal%20User%20Representations%20from%20Multiple%20E-commerce%20Tasks.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (SIGIR) [ESMM] Entire Space Multi-Task Model - An Effective Approach for Estimating Post-Click Conversion Rate](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2018%20%28Alibaba%29%20%28SIGIR%29%20%5BESMM%5D%20Entire%20Space%20Multi-Task%20Model%20-%20An%20Effective%20Approach%20for%20Estimating%20Post-Click%20Conversion%20Rate.pdf) \u003Cbr \u002F>\n* [2018 (CVPR) Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2018%20%28CVPR%29%20Multi-Task%20Learning%20Using%20Uncertainty%20to%20Weigh%20Losses%20for%20Scene%20Geometry%20and%20Semantics.pdf) \u003Cbr \u002F>\n* [2018 (Google) (KDD) ** [MMoE] Modeling task relationships in multi-task learning with multi-gate mixture-of-experts](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2018%20%28Google%29%20%28KDD%29%20%2A%2A%20%5BMMoE%5D%20Modeling%20task%20relationships%20in%20multi-task%20learning%20with%20multi-gate%20mixture-of-experts.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (CIKM) Multi-task based Sales Predictions for Online Promotions](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28Alibaba%29%20%28CIKM%29%20Multi-task%20based%20Sales%20Predictions%20for%20Online%20Promotions.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (Recys) A Pareto-Eficient Algorithm for Multiple Objective Optimization in E-Commerce Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28Alibaba%29%20%28Recys%29%20A%20Pareto-Eficient%20Algorithm%20for%20Multiple%20Objective%20Optimization%20in%20E-Commerce%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019 (Google) (AAAI) SNR Sub-Network Routing for Flexible Parameter Sharing in Multi-Task Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28Google%29%20%28AAAI%29%20SNR%20Sub-Network%20Routing%20for%20Flexible%20Parameter%20Sharing%20in%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2019 (Google) (Recsys) ** [Youtube Multi-task] Recommending what video to watch next - a multitask ranking system](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28Google%29%20%28Recsys%29%20%2A%2A%20%5BYoutube%20Multi-task%5D%20Recommending%20what%20video%20to%20watch%20next%20-%20a%20multitask%20ranking%20system.pdf) \u003Cbr \u002F>\n* [2019 (NIPS) Pareto Multi-Task Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28NIPS%29%20Pareto%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (SIGIR) [ESM2] Entire Space Multi-Task Modeling via Post-Click Behavior Decomposition for Conversion Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Alibaba%29%20%28SIGIR%29%20%5BESM2%5D%20Entire%20Space%20Multi-Task%20Modeling%20via%20Post-Click%20Behavior%20Decomposition%20for%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (WWW) Large-scale Causal Approaches to Debiasing Post-click Conversion Rate Estimation with Multi-task Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Alibaba%29%20%28WWW%29%20Large-scale%20Causal%20Approaches%20to%20Debiasing%20Post-click%20Conversion%20Rate%20Estimation%20with%20Multi-task%20Learning.pdf) \u003Cbr \u002F>\n* [2020 (Amazon) (WWW) Multi-Objective Ranking Optimization for Product Search Using Stochastic Label Aggregation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Amazon%29%20%28WWW%29%20Multi-Objective%20Ranking%20Optimization%20for%20Product%20Search%20Using%20Stochastic%20Label%20Aggregation.pdf) \u003Cbr \u002F>\n* [2020 (Google) (KDD) [MoSE] Multitask Mixture of Sequential Experts for User Activity Streams](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Google%29%20%28KDD%29%20%5BMoSE%5D%20Multitask%20Mixture%20of%20Sequential%20Experts%20for%20User%20Activity%20Streams.pdf) \u003Cbr \u002F>\n* [2020 (JD) (CIKM) *[DMT] Deep Multifaceted Transformers for Multi-objective Ranking in Large-Scale E-commerce Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28JD%29%20%28CIKM%29%20%2A%5BDMT%5D%20Deep%20Multifaceted%20Transformers%20for%20Multi-objective%20Ranking%20in%20Large-Scale%20E-commerce%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2020 (Tencent) (Recsys) **  [PLE] Progressive Layered Extraction (PLE) - A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Tencent%29%20%28Recsys%29%20%2A%2A%20%20%5BPLE%5D%20Progressive%20Layered%20Extraction%20%28PLE%29%20-%20A%20Novel%20Multi-Task%20Learning%20%28MTL%29%20Model%20for%20Personalized%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (SIGIR) [HM3] Hierarchically Modeling Micro and Macro Behaviors via Multi-Task Learning for Conversion Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Alibaba%29%20%28SIGIR%29%20%5BHM3%5D%20Hierarchically%20Modeling%20Micro%20and%20Macro%20Behaviors%20via%20Multi-Task%20Learning%20for%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (SIGIR) [MSSM] MSSM - A Multiple-level Sparse Sharing Model for Efficient Multi-Task Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Alibaba%29%20%28SIGIR%29%20%5BMSSM%5D%20MSSM%20-%20A%20Multiple-level%20Sparse%20Sharing%20Model%20for%20Efficient%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2021 (Baidu) (SIGIR) [GemNN] GemNN - Gating-Enhanced Multi-Task Neural Networks with Feature Interaction Learning for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Baidu%29%20%28SIGIR%29%20%5BGemNN%5D%20GemNN%20-%20Gating-Enhanced%20Multi-Task%20Neural%20Networks%20with%20Feature%20Interaction%20Learning%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Google) (Arxiv) [DSelect-k] DSelect-k Differentiable Selection in the Mixture of Experts with Applications to Multi-Task Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Google%29%20%28Arxiv%29%20%5BDSelect-k%5D%20DSelect-k%20Differentiable%20Selection%20in%20the%20Mixture%20of%20Experts%20with%20Applications%20to%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2021 (Google) (KDD) Understanding and Improving Fairness-Accuracy Trade-offs in Multi-Task Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Google%29%20%28KDD%29%20Understanding%20and%20Improving%20Fairness-Accuracy%20Trade-offs%20in%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2021 (JD) (ICDE) Adversarial Mixture Of Experts with Category Hierarchy Soft Constraint](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28JD%29%20%28ICDE%29%20Adversarial%20Mixture%20Of%20Experts%20with%20Category%20Hierarchy%20Soft%20Constraint.pdf) \u003Cbr \u002F>\n* [2021 (Meituan) (KDD) Modeling the Sequential Dependence among Audience Multi-step Conversions with Multi-task Learning in Targeted Display Advertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Meituan%29%20%28KDD%29%20Modeling%20the%20Sequential%20Dependence%20among%20Audience%20Multi-step%20Conversions%20with%20Multi-task%20Learning%20in%20Targeted%20Display%20Advertising.pdf) \u003Cbr \u002F>\n* [2021 (Tencent) (Arxiv) Mixture of Virtual-Kernel Experts for Multi-Objective User Profile Modeling](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Tencent%29%20%28Arxiv%29%20Mixture%20of%20Virtual-Kernel%20Experts%20for%20Multi-Objective%20User%20Profile%20Modeling.pdf) \u003Cbr \u002F>\n* [2021 (Tencent) (WWW) Personalized Approximate Pareto-Efficient Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Tencent%29%20%28WWW%29%20Personalized%20Approximate%20Pareto-Efficient%20Recommendation.pdf) \u003Cbr \u002F>\n* [2022 (Google) (WWW) Can Small Heads Help? Understanding and Improving Multi-Task Generalization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2022%20%28Google%29%20%28WWW%29%20Can%20Small%20Heads%20Help%3F%20Understanding%20and%20Improving%20Multi-Task%20Generalization.pdf) \u003Cbr \u002F>\n* [2023 (Airbnb) (KDD) Optimizing Airbnb Search Journey with Multi-task Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Airbnb%29%20%28KDD%29%20Optimizing%20Airbnb%20Search%20Journey%20with%20Multi-task%20Learning.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (CIKM) [DTRN] Deep Task-specific Bottom Representation Network for Multi-Task Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BDTRN%5D%20Deep%20Task-specific%20Bottom%20Representation%20Network%20for%20Multi-Task%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023 (Google) (CIKM) Multitask Ranking System for Immersive Feed and No More Clicks - A Case Study of Short-Form Video Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Google%29%20%28CIKM%29%20Multitask%20Ranking%20System%20for%20Immersive%20Feed%20and%20No%20More%20Clicks%20-%20A%20Case%20Study%20of%20Short-Form%20Video%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023 (Google) (KDD) Improving Training Stability for Multitask Ranking Models in Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Google%29%20%28KDD%29%20Improving%20Training%20Stability%20for%20Multitask%20Ranking%20Models%20in%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2023 (Meta) (KDD) AdaTT - Adaptive Task-to-Task Fusion Network for Multitask Learning in Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Meta%29%20%28KDD%29%20AdaTT%20-%20Adaptive%20Task-to-Task%20Fusion%20Network%20for%20Multitask%20Learning%20in%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024 (Airbnb) (KDD) Multi-objective Learning to Rank by Model Distillation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Airbnb%29%20%28KDD%29%20Multi-objective%20Learning%20to%20Rank%20by%20Model%20Distillation.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (KDD) [GradCraft] GradCraft - Elevating Multi-task Recommendations through Holistic Gradient Crafting](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Kuaishou%29%20%28KDD%29%20%5BGradCraft%5D%20GradCraft%20-%20Elevating%20Multi-task%20Recommendations%20through%20Holistic%20Gradient%20Crafting.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) [HoME] HoME - Hierarchy of Multi-Gate Experts for Multi-Task Learning at Kuaishou](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Kuaishou%29%20%5BHoME%5D%20HoME%20-%20Hierarchy%20of%20Multi-Gate%20Experts%20for%20Multi-Task%20Learning%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2024 (Shopee) (KDD) [ResFlow] Residual Multi-Task Learner for Applied Ranking](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Shopee%29%20%28KDD%29%20%5BResFlow%5D%20Residual%20Multi-Task%20Learner%20for%20Applied%20Ranking.pdf) \u003Cbr \u002F>\n* [2024 (Tencent) (KDD) [STEM] Ads Recommendation in a Collapsed and Entangled World](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Tencent%29%20%28KDD%29%20%5BSTEM%5D%20Ads%20Recommendation%20in%20a%20Collapsed%20and%20Entangled%20World.pdf) \u003Cbr \u002F>\n* [2025 (Baidu) (KDD) [RankExpert] RankExpert - A Mixture of Textual-and-Behavioral Experts for Multi-Objective Learning-to-Rank in Web Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2025%20%28Baidu%29%20%28KDD%29%20%5BRankExpert%5D%20RankExpert%20-%20A%20Mixture%20of%20Textual-and-Behavioral%20Experts%20for%20Multi-Objective%20Learning-to-Rank%20in%20Web%20Search.pdf) \u003Cbr \u002F>\n* [2025 (Byetedance) (CIKM) [PMTA] PMTA - Perception-Aware Multi-Task Transformer Network for Personalized Multi-Domain Adaptation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2025%20%28Byetedance%29%20%28CIKM%29%20%5BPMTA%5D%20PMTA%20-%20Perception-Aware%20Multi-Task%20Transformer%20Network%20for%20Personalized%20Multi-Domain%20Adaptation.pdf) \u003Cbr \u002F>\n* [2025 （Alibaba) (CIKM) [MAL] See Beyond a Single View - Multi-Attribution Learning Leads to Better Conversion Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2025%20%EF%BC%88Alibaba%29%20%28CIKM%29%20%5BMAL%5D%20See%20Beyond%20a%20Single%20View%20-%20Multi-Attribution%20Learning%20Leads%20to%20Better%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n\n#### ParameterServer\n* [2014 (Baidu) (OSDI) Scaling Distributed Machine Learning with the Parameter Server](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FParameterServer\u002F2014%20%28Baidu%29%20%28OSDI%29%20Scaling%20Distributed%20Machine%20Learning%20with%20the%20Parameter%20Server.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (DLP-KDD) [XDL] XDL - An Industrial Deep Learning Framework for High-dimensional Sparse Data](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FParameterServer\u002F2019%20%28Alibaba%29%20%28DLP-KDD%29%20%5BXDL%5D%20XDL%20-%20An%20Industrial%20Deep%20Learning%20Framework%20for%20High-dimensional%20Sparse%20Data.pdf) \u003Cbr \u002F>\n* [2020 (Bytedance) (OSDI) [BytePS] A Unified Architecture for Accelerating Distributed DNN Training in Heterogeneous GPU:CPU Clusters](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FParameterServer\u002F2020%20%28Bytedance%29%20%28OSDI%29%20%5BBytePS%5D%20A%20Unified%20Architecture%20for%20Accelerating%20Distributed%20DNN%20Training%20in%20Heterogeneous%20GPU%3ACPU%20Clusters.pdf) \u003Cbr \u002F>\n* [2022 (Kuaishou) (KDD) [Persia] Persia - An Open, Hybrid System Scaling Deep Learning-based Recommenders up to 100 Trillion Parameters](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FParameterServer\u002F2022%20%28Kuaishou%29%20%28KDD%29%20%5BPersia%5D%20Persia%20-%20An%20Open%2C%20Hybrid%20System%20Scaling%20Deep%20Learning-based%20Recommenders%20up%20to%20100%20Trillion%20Parameters.pdf) \u003Cbr \u002F>\n\n#### Pre-training\n* [2019 (Alibaba) (IJCAI) [DeepMCP] Representation Learning-Assisted Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FPre-training\u002F2019%20%28Alibaba%29%20%28IJCAI%29%20%5BDeepMCP%5D%20Representation%20Learning-Assisted%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (SIGIR) [BERT4Rec] (Alibaba) (SIGIR2019) BERT4Rec - Sequential Recommendation with Bidirectional Encoder Representations from Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FPre-training\u002F2019%20%28SIGIR%29%20%5BBERT4Rec%5D%20%28Alibaba%29%20%28SIGIR2019%29%20BERT4Rec%20-%20Sequential%20Recommendation%20with%20Bidirectional%20Encoder%20Representations%20from%20Transformer.pdf) \u003Cbr \u002F>\n\n#### Sequence-Modeling\n* [2016 (Google) (RecSys) **[Youtube DNN] Deep Neural Networks for YouTube Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2016%20%28Google%29%20%28RecSys%29%20%2A%2A%5BYoutube%20DNN%5D%20Deep%20Neural%20Networks%20for%20YouTube%20Recommendations.pdf) \u003Cbr \u002F>\n* [2017 (Google) (NIPS) ** Attention Is All You Need](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2017%20%28Google%29%20%28NIPS%29%20%2A%2A%20Attention%20Is%20All%20You%20Need.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (KDD) **[DIN] Deep Interest Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2018%20%28Alibaba%29%20%28KDD%29%20%2A%2A%5BDIN%5D%20Deep%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (KDD) [DUPN] Perceive Your Users in Depth - Learning Universal User Representations from Multiple E-commerce Tasks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2018%20%28Alibaba%29%20%28KDD%29%20%5BDUPN%5D%20Perceive%20Your%20Users%20in%20Depth%20-%20Learning%20Universal%20User%20Representations%20from%20Multiple%20E-commerce%20Tasks.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (AAAI) **[DIEN] Deep Interest Evolution Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28AAAI%29%20%2A%2A%5BDIEN%5D%20Deep%20Interest%20Evolution%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (IJCAI) [DSIN] Deep Session Interest Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28IJCAI%29%20%5BDSIN%5D%20Deep%20Session%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) [BST] Behavior Sequence Transformer for E-commerce Recommendation in Alibaba](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BBST%5D%20Behavior%20Sequence%20Transformer%20for%20E-commerce%20Recommendation%20in%20Alibaba.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) [DSTN] Deep Spatio-Temporal Neural Networks for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BDSTN%5D%20Deep%20Spatio-Temporal%20Neural%20Networks%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (WWW) [TiSSA] TiSSA - A Time Slice Self-Attention Approach for Modeling Sequential User Behaviors](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28WWW%29%20%5BTiSSA%5D%20TiSSA%20-%20A%20Time%20Slice%20Self-Attention%20Approach%20for%20Modeling%20Sequential%20User%20Behaviors.pdf) \u003Cbr \u002F>\n* [2019 (Tencent) (KDD) [RALM] TReal-time Attention Based Look-alike Model for Recommender System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Tencent%29%20%28KDD%29%20%5BRALM%5D%20TReal-time%20Attention%20Based%20Look-alike%20Model%20for%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (SIGIR) [DHAN] Deep Interest with Hierarchical Attention Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28Alibaba%29%20%28SIGIR%29%20%5BDHAN%5D%20Deep%20Interest%20with%20Hierarchical%20Attention%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020 (Google) (KDD) [Google Drive] Improving Recommendation Quality in Google Drive](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28Google%29%20%28KDD%29%20%5BGoogle%20Drive%5D%20Improving%20Recommendation%20Quality%20in%20Google%20Drive.pdf) \u003Cbr \u002F>\n* [2020 (JD) (CIKM) **[DMT] Deep Multifaceted Transformers for Multi-objective Ranking in Large-Scale E-commerce Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28JD%29%20%28CIKM%29%20%2A%2A%5BDMT%5D%20Deep%20Multifaceted%20Transformers%20for%20Multi-objective%20Ranking%20in%20Large-Scale%20E-commerce%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2020 (JD) (NIPS) [KFAtt] Kalman Filtering Attention for User Behavior Modeling in CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28JD%29%20%28NIPS%29%20%5BKFAtt%5D%20Kalman%20Filtering%20Attention%20for%20User%20Behavior%20Modeling%20in%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2020 (JD) (WSDM) [HUP] Hierarchical User Profiling for E-commerce Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28JD%29%20%28WSDM%29%20%5BHUP%5D%20Hierarchical%20User%20Profiling%20for%20E-commerce%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (WSDM) [RACP] Modeling Users’ Contextualized Page-wise Feedback for Click-Through Rate Prediction in E-commerce Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2022%20%28Alibaba%29%20%28WSDM%29%20%5BRACP%5D%20Modeling%20Users%E2%80%99%20Contextualized%20Page-wise%20Feedback%20for%20Click-Through%20Rate%20Prediction%20in%20E-commerce%20Search.pdf) \u003Cbr \u002F>\n* [2022 (JD) (WWW) Implicit User Awareness Modeling via Candidate Items for CTR Prediction in Search Ads](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2022%20%28JD%29%20%28WWW%29%20Implicit%20User%20Awareness%20Modeling%20via%20Candidate%20Items%20for%20CTR%20Prediction%20in%20Search%20Ads.pdf) \u003Cbr \u002F>\n* [2022 (WWW) [FMLP] Filter-enhanced MLP is All You Need for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2022%20%28WWW%29%20%5BFMLP%5D%20Filter-enhanced%20MLP%20is%20All%20You%20Need%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023 (JD) (CIKM) [IUI] IUI - Intent-Enhanced User Interest Modeling for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2023%20%28JD%29%20%28CIKM%29%20%5BIUI%5D%20IUI%20-%20Intent-Enhanced%20User%20Interest%20Modeling%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Meituan) (CIKM) [DCIN] Deep Context Interest Network for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2023%20%28Meituan%29%20%28CIKM%29%20%5BDCIN%5D%20Deep%20Context%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Pinterest) (KDD) TransAct - Transformer-based Realtime User Action Model for Recommendation at Pinterest](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2023%20%28Pinterest%29%20%28KDD%29%20TransAct%20-%20Transformer-based%20Realtime%20User%20Action%20Model%20for%20Recommendation%20at%20Pinterest.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (SIGIR) [FIM] FIM - Frequency-Aware Multi-View Interest Modeling for Local-Life Service Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2025%20%28Kuaishou%29%20%28SIGIR%29%20%5BFIM%5D%20FIM%20-%20Frequency-Aware%20Multi-View%20Interest%20Modeling%20for%20Local-Life%20Service%20Recommendation.pdf) \u003Cbr \u002F>\n\n#### Sequence-Modeling-Longterm\n* [2019 (Alibaba) (KDD) [MIMN] Practice on Long Sequential User Behavior Modeling for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BMIMN%5D%20Practice%20on%20Long%20Sequential%20User%20Behavior%20Modeling%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019 (Google) (WWW) Towards Neural Mixture Recommender for Long Range Dependent User Sequences](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2019%20%28Google%29%20%28WWW%29%20Towards%20Neural%20Mixture%20Recommender%20for%20Long%20Range%20Dependent%20User%20Sequences.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (Arxiv) ** [SIM] Search-based User Interest Modeling with Lifelong Sequential Behavior Data for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2020%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BSIM%5D%20Search-based%20User%20Interest%20Modeling%20with%20Lifelong%20Sequential%20Behavior%20Data%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020 (ICLR) Reformer - The Efficient Transformer ](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2020%20%28ICLR%29%20Reformer%20-%20The%20Efficient%20Transformer%20.pdf) \u003Cbr \u002F>\n* [2020 (SIGIR) [UBR4CTR] User Behavior Retrieval for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2020%20%28SIGIR%29%20%5BUBR4CTR%5D%20User%20Behavior%20Retrieval%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (Arxiv) [ETA] End-to-End User Behavior Retrieval in Click-Through Rate Prediction Model](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2021%20%28Alibaba%29%20%28Arxiv%29%20%5BETA%5D%20End-to-End%20User%20Behavior%20Retrieval%20in%20Click-Through%20Rate%20Prediction%20Model.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (Arxiv) ** [ETA] Efficient Long Sequential User Data Modeling for Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2022%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BETA%5D%20Efficient%20Long%20Sequential%20User%20Data%20Modeling%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2022 (Meituan) (CIKM) [SDIM] Sampling Is All You Need on Modeling Long-Term User Behaviors for CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2022%20%28Meituan%29%20%28CIKM%29%20%5BSDIM%5D%20Sampling%20Is%20All%20You%20Need%20on%20Modeling%20Long-Term%20User%20Behaviors%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Kuaishou) (Arixiv) [TWIN] TWIN - TWo-stage Interest Network for Lifelong User Behavior Modeling in CTR Prediction at Kuaishou](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2023%20%28Kuaishou%29%20%28Arixiv%29%20%5BTWIN%5D%20TWIN%20-%20TWo-stage%20Interest%20Network%20for%20Lifelong%20User%20Behavior%20Modeling%20in%20CTR%20Prediction%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2023 (Kuaishou) (CIKM) [QIN] Query-dominant User Interest Network for Large-Scale Search Ranking](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2023%20%28Kuaishou%29%20%28CIKM%29%20%5BQIN%5D%20Query-dominant%20User%20Interest%20Network%20for%20Large-Scale%20Search%20Ranking.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (CIKM) [TWINv2] TWIN V2 - Scaling Ultra-Long User Behavior Sequence Modeling for Enhanced CTR Prediction at Kuaishou](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2024%20%28Kuaishou%29%20%28CIKM%29%20%5BTWINv2%5D%20TWIN%20V2%20-%20Scaling%20Ultra-Long%20User%20Behavior%20Sequence%20Modeling%20for%20Enhanced%20CTR%20Prediction%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2024 (Tencent) (KDD) [LCN] Cross-Domain LifeLong Sequential Modeling for Online Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2024%20%28Tencent%29%20%28KDD%29%20%5BLCN%5D%20Cross-Domain%20LifeLong%20Sequential%20Modeling%20for%20Online%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [MUSE] MUSE - A Simple Yet Effective Multimodal Search-Based Framework for Lifelong User Interest Modeling](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BMUSE%5D%20MUSE%20-%20A%20Simple%20Yet%20Effective%20Multimodal%20Search-Based%20Framework%20for%20Lifelong%20User%20Interest%20Modeling.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (Arxiv) [LONGER] LONGER - Scaling Up Long Sequence Modeling in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BLONGER%5D%20LONGER%20-%20Scaling%20Up%20Long%20Sequence%20Modeling%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (KDD) [HiT-LBM] Hierarchical Tree Search-based User Lifelong Behavior Modeling on Large Language Model](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Kuaishou%29%20%28KDD%29%20%5BHiT-LBM%5D%20Hierarchical%20Tree%20Search-based%20User%20Lifelong%20Behavior%20Modeling%20on%20Large%20Language%20Model.pdf) \u003Cbr \u002F>\n* [2025 (Meta) (Arxiv) [VISTA] Massive Memorization with Hundreds of Trillions of Parameters for Sequential Transducer Generative Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Meta%29%20%28Arxiv%29%20%5BVISTA%5D%20Massive%20Memorization%20with%20Hundreds%20of%20Trillions%20of%20Parameters%20for%20Sequential%20Transducer%20Generative%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Meta) (KDD) DV365 - Extremely Long User History Modeling at Instagram](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Meta%29%20%28KDD%29%20DV365%20-%20Extremely%20Long%20User%20History%20Modeling%20at%20Instagram.pdf) \u003Cbr \u002F>\n* [2025 (Pinterest) (Arxiv) [TransActV2]TransAct V2 - Lifelong User Action Sequence Modeling on Pinterest Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Pinterest%29%20%28Arxiv%29%20%5BTransActV2%5DTransAct%20V2%20-%20Lifelong%20User%20Action%20Sequence%20Modeling%20on%20Pinterest%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 （Bytedance) ** (Arxiv) [STCA] Make It Long, Keep It Fast - End-to-End 10k-Sequence Modeling at Billion Scale on Douyin](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%EF%BC%88Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BSTCA%5D%20Make%20It%20Long%2C%20Keep%20It%20Fast%20-%20End-to-End%2010k-Sequence%20Modeling%20at%20Billion%20Scale%20on%20Douyin.pdf) \u003Cbr \u002F>\n\n#### Transfer_Learning\n* [2014 (Google) (NIPS) [Knoledge Distillation] Distilling the Knowledge in a Neural Network](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002F2014%20%28Google%29%20%28NIPS%29%20%5BKnoledge%20Distillation%5D%20Distilling%20the%20Knowledge%20in%20a%20Neural%20Network.pdf) \u003Cbr \u002F>\n* [2015 (ICLR) [Fitnets] Fitnets - Hints for thin deep nets](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002F2015%20%28ICLR%29%20%5BFitnets%5D%20Fitnets%20-%20Hints%20for%20thin%20deep%20nets.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (AAAI) [Rocket] Rocket launching - A universal and efficient framework for training well-performing light net](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002F2018%20%28Alibaba%29%20%28AAAI%29%20%5BRocket%5D%20Rocket%20launching%20-%20A%20universal%20and%20efficient%20framework%20for%20training%20well-performing%20light%20net.pdf) \u003Cbr \u002F>\n* [2018 (KDD)[Ranking Distillation] Ranking distillation - Learning compact ranking models with high performance for recommender system](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002F2018%20%28KDD%29%5BRanking%20Distillation%5D%20Ranking%20distillation%20-%20Learning%20compact%20ranking%20models%20with%20high%20performance%20for%20recommender%20system.pdf) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002FCross-domain) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002FTransfer) \u003Cbr \u002F>\n\n#### Trigger\n* [2022 (Alibaba) (WWW) Deep Interest Highlight Network for Click-Through Rate Prediction in Trigger-Induced Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTrigger\u002F2022%20%28Alibaba%29%20%28WWW%29%20Deep%20Interest%20Highlight%20Network%20for%20Click-Through%20Rate%20Prediction%20in%20Trigger-Induced%20Recommendation.pdf) \u003Cbr \u002F>\n\n## 04_Post-ranking\n* [1998 (SIGIR) **  [MRR] The Use of MMR, Diversity-Based Reranking for Reordering Documents and Producing Summaries](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F1998%20%28SIGIR%29%20%2A%2A%20%20%5BMRR%5D%20The%20Use%20of%20MMR%2C%20Diversity-Based%20Reranking%20for%20Reordering%20Documents%20and%20Producing%20Summaries.pdf) \u003Cbr \u002F>\n* [2005 (WWW) Improving Recommendation Lists Through Topic Diversification](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2005%20%28WWW%29%20Improving%20Recommendation%20Lists%20Through%20Topic%20Diversification.pdf) \u003Cbr \u002F>\n* [2008 (SIGIR) [α-NDCG] Novelty and Diversity in Information Retrieval Evaluation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2008%20%28SIGIR%29%20%5B%CE%B1-NDCG%5D%20Novelty%20and%20Diversity%20in%20Information%20Retrieval%20Evaluation.pdf) \u003Cbr \u002F>\n* [2009 (Microsoft) (WSDM) Diversifying Search Results](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2009%20%28Microsoft%29%20%28WSDM%29%20Diversifying%20Search%20Results.pdf) \u003Cbr \u002F>\n* [2010 (WWW) Exploiting Query Reformulations for Web Search Result Diversification](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2010%20%28WWW%29%20Exploiting%20Query%20Reformulations%20for%20Web%20Search%20Result%20Diversification.pdf) \u003Cbr \u002F>\n* [2016 (Amazon) (RecSys) Adaptive, Personalized Diversity for Visual Discovery](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2016%20%28Amazon%29%20%28RecSys%29%20Adaptive%2C%20Personalized%20Diversity%20for%20Visual%20Discovery.pdf) \u003Cbr \u002F>\n* [2017 (Hulu) (NIPS) [DPP] Fast Greedy MAP Inference for Determinantal Point Process to Improve Recommendation Diversity](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2017%20%28Hulu%29%20%28NIPS%29%20%5BDPP%5D%20Fast%20Greedy%20MAP%20Inference%20for%20Determinantal%20Point%20Process%20to%20Improve%20Recommendation%20Diversity.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (IJCAI) [Alibaba GMV] Globally Optimized Mutual Influence Aware Ranking in E-Commerce Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2018%20%28Alibaba%29%20%28IJCAI%29%20%5BAlibaba%20GMV%5D%20Globally%20Optimized%20Mutual%20Influence%20Aware%20Ranking%20in%20E-Commerce%20Search.pdf) \u003Cbr \u002F>\n* [2018 (Google) (CIKM) [DPP] Practical Diversified Recommendations on YouTube with Determinantal Point Processes](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2018%20%28Google%29%20%28CIKM%29%20%5BDPP%5D%20Practical%20Diversified%20Recommendations%20on%20YouTube%20with%20Determinantal%20Point%20Processes.pdf) \u003Cbr \u002F>\n* [2018 (SIGIR) [DLCM] Learning a Deep Listwise Context Model for Ranking Refinement](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2018%20%28SIGIR%29%20%5BDLCM%5D%20Learning%20a%20Deep%20Listwise%20Context%20Model%20for%20Ranking%20Refinement.pdf) \u003Cbr \u002F>\n* [2019  (Alibaba) (WWW) [Value-based RL] Value-aware Recommendation based on Reinforcement Profit Maximization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%20%28Alibaba%29%20%28WWW%29%20%5BValue-based%20RL%5D%20Value-aware%20Recommendation%20based%20on%20Reinforcement%20Profit%20Maximization.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) [GAttN] Exact-K Recommendation via Maximal Clique Optimization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BGAttN%5D%20Exact-K%20Recommendation%20via%20Maximal%20Clique%20Optimization.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (RecSys)  ** [PRM] Personalized Re-ranking for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Alibaba%29%20%28RecSys%29%20%20%2A%2A%20%5BPRM%5D%20Personalized%20Re-ranking%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019 (Google) (Arxiv) Reinforcement Learning for Slate-based Recommender Systems - A Tractable Decomposition and Practical Methodology](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Google%29%20%28Arxiv%29%20Reinforcement%20Learning%20for%20Slate-based%20Recommender%20Systems%20-%20A%20Tractable%20Decomposition%20and%20Practical%20Methodology.pdf) \u003Cbr \u002F>\n* [2019 (Google) (Arxiv) Seq2slate - Re-ranking and slate optimization with rnns](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Google%29%20%28Arxiv%29%20Seq2slate%20-%20Re-ranking%20and%20slate%20optimization%20with%20rnns.pdf) \u003Cbr \u002F>\n* [2019 (Google) (IJCAI) [SlateQ] SLATEQ - A Tractable Decomposition for Reinforcement Learning with Recommendation Sets](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Google%29%20%28IJCAI%29%20%5BSlateQ%5D%20SLATEQ%20-%20A%20Tractable%20Decomposition%20for%20Reinforcement%20Learning%20with%20Recommendation%20Sets.pdf) \u003Cbr \u002F>\n* [2020 (Airbnb) (KDD) Managing Diversity in Airbnb Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2020%20%28Airbnb%29%20%28KDD%29%20Managing%20Diversity%20in%20Airbnb%20Search.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (CIKM) [EdgeRec] EdgeRec - Recommender System on Edge in Mobile Taobao](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2020%20%28Alibaba%29%20%28CIKM%29%20%5BEdgeRec%5D%20EdgeRec%20-%20Recommender%20System%20on%20Edge%20in%20Mobile%20Taobao.pdf) \u003Cbr \u002F>\n* [2020 (Huawei) (Arxiv) Personalized Re-ranking for Improving Diversity in Live Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2020%20%28Huawei%29%20%28Arxiv%29%20Personalized%20Re-ranking%20for%20Improving%20Diversity%20in%20Live%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (Arxiv) **  [PRS] Revisit Recommender System in the Permutation Prospective](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2021%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%20%5BPRS%5D%20Revisit%20Recommender%20System%20in%20the%20Permutation%20Prospective.pdf) \u003Cbr \u002F>\n* [2021 (Google) (WSDM) User Response Models to Improve a REINFORCE Recommender System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2021%20%28Google%29%20%28WSDM%29%20User%20Response%20Models%20to%20Improve%20a%20REINFORCE%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2021 (Microsoft) Diversity on the Go! Streaming Determinantal Point Processes under a Maximum Induced Cardinality Objective](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2021%20%28Microsoft%29%20Diversity%20on%20the%20Go%21%20Streaming%20Determinantal%20Point%20Processes%20under%20a%20Maximum%20Induced%20Cardinality%20Objective.pdf) \u003Cbr \u002F>\n* [2023 (Amazon) (KDD) RankFormer - Listwise Learning-to-Rank Using Listwide Labels](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2023%20%28Amazon%29%20%28KDD%29%20RankFormer%20-%20Listwise%20Learning-to-Rank%20Using%20Listwide%20Labels.pdf) \u003Cbr \u002F>\n* [2023 (Meituan) (KDD) PIER - Permutation-Level Interest-Based End-to-End Re-ranking Framework in E-commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2023%20%28Meituan%29%20%28KDD%29%20PIER%20-%20Permutation-Level%20Interest-Based%20End-to-End%20Re-ranking%20Framework%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (KDD) [NAR4Rec] Non-autoregressive Generative Models for Reranking Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2024%20%28Kuaishou%29%20%28KDD%29%20%5BNAR4Rec%5D%20Non-autoregressive%20Generative%20Models%20for%20Reranking%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (SIGIR) [SORT-Gen] A Generative Re-ranking Model for List-level Multi-objective Optimization at Taobao](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2025%20%28Alibaba%29%20%28SIGIR%29%20%5BSORT-Gen%5D%20A%20Generative%20Re-ranking%20Model%20for%20List-level%20Multi-objective%20Optimization%20at%20Taobao.pdf) \u003Cbr \u002F>\n\n#### Seq2Slate\n* [2015 (Google) (Arxiv)  Deep Reinforcement Learning in Large Discrete Action Spaces ](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2015%20%28Google%29%20%28Arxiv%29%20%20Deep%20Reinforcement%20Learning%20in%20Large%20Discrete%20Action%20Spaces%20.pdf) \u003Cbr \u002F>\n* [2015 (Google) (Arxiv) Deep Reinforcement Learning with Attention for Slate Markov Decision Processes with High-Dimensional States and Actions](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2015%20%28Google%29%20%28Arxiv%29%20Deep%20Reinforcement%20Learning%20with%20Attention%20for%20Slate%20Markov%20Decision%20Processes%20with%20High-Dimensional%20States%20and%20Actions.pdf) \u003Cbr \u002F>\n* [2017 (KDD) [DCM] Deep Choice Model Using Pointer Networks for Airline Itinerary Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2017%20%28KDD%29%20%5BDCM%5D%20Deep%20Choice%20Model%20Using%20Pointer%20Networks%20for%20Airline%20Itinerary%20Prediction.pdf) \u003Cbr \u002F>\n* [2018 (Microsoft) (EMNLP) [RL4NMT] A study of reinforcement learning for neural machine translation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2018%20%28Microsoft%29%20%28EMNLP%29%20%5BRL4NMT%5D%20A%20study%20of%20reinforcement%20learning%20for%20neural%20machine%20translation.pdf) \u003Cbr \u002F>\n* [2019 (Google) (Arxiv) Seq2slate - Re-ranking and slate optimization with rnns](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2019%20%28Google%29%20%28Arxiv%29%20Seq2slate%20-%20Re-ranking%20and%20slate%20optimization%20with%20rnns.pdf) \u003Cbr \u002F>\n\n## 05_Relevance-ranking\n* [2013 (Microsoft) (CIKM) [DSSM] Learning deep structured semantic models for web search using clickthrough data](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2013%20%28Microsoft%29%20%28CIKM%29%20%5BDSSM%5D%20Learning%20deep%20structured%20semantic%20models%20for%20web%20search%20using%20clickthrough%20data.pdf) \u003Cbr \u002F>\n* [2016 (Yahoo) (KDD) Ranking Relevance in Yahoo Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2016%20%28Yahoo%29%20%28KDD%29%20Ranking%20Relevance%20in%20Yahoo%20Search.pdf) \u003Cbr \u002F>\n* [2020 (ICLR) [StructBERT] StructBERT - Incorporating Language Structures into Pre-training for Deep Language Understanding](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2020%20%28ICLR%29%20%5BStructBERT%5D%20StructBERT%20-%20Incorporating%20Language%20Structures%20into%20Pre-training%20for%20Deep%20Language%20Understanding.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (WWW) [MASM] Learning a Product Relevance Model from Click-Through Data in E-Commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2021%20%28Alibaba%29%20%28WWW%29%20%5BMASM%5D%20Learning%20a%20Product%20Relevance%20Model%20from%20Click-Through%20Data%20in%20E-Commerce.pdf) \u003Cbr \u002F>\n* [2022 (Alibaba) (KDD) [ReprBERT] ReprBERT - Distilling BERT to an Efficient Representation-Based Relevance Model for E-Commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2022%20%28Alibaba%29%20%28KDD%29%20%5BReprBERT%5D%20ReprBERT%20-%20Distilling%20BERT%20to%20an%20Efficient%20Representation-Based%20Relevance%20Model%20for%20E-Commerce.pdf) \u003Cbr \u002F>\n* [2023 (Meituan) (CIKM) [SPM] SPM - Structured Pretraining and Matching Architectures for Relevance Modeling in Meituan Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2023%20%28Meituan%29%20%28CIKM%29%20%5BSPM%5D%20SPM%20-%20Structured%20Pretraining%20and%20Matching%20Architectures%20for%20Relevance%20Modeling%20in%20Meituan%20Search.pdf) \u003Cbr \u002F>\n* [2024 (Alibaba) (KDD) [DeepBoW] Deep Bag-of-Words Model - An Efficient and Interpretable Relevance Architecture for Chinese E-Commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2024%20%28Alibaba%29%20%28KDD%29%20%5BDeepBoW%5D%20Deep%20Bag-of-Words%20Model%20-%20An%20Efficient%20and%20Interpretable%20Relevance%20Architecture%20for%20Chinese%20E-Commerce.pdf) \u003Cbr \u002F>\n* [2024 （Walmart) (SIGIR) Large Language Models for Relevance Judgment in Product Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2024%20%EF%BC%88Walmart%29%20%28SIGIR%29%20Large%20Language%20Models%20for%20Relevance%20Judgment%20in%20Product%20Search.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) LORE - A Large Generative Model for Search Relevance](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20LORE%20-%20A%20Large%20Generative%20Model%20for%20Search%20Relevance.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (WWW) [ELLM] Explainable LLM-driven Multi-dimensional Distillation for E-Commerce Relevance Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%28Alibaba%29%20%28WWW%29%20%5BELLM%5D%20Explainable%20LLM-driven%20Multi-dimensional%20Distillation%20for%20E-Commerce%20Relevance%20Learning.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [HCMRM] HCMRM -A High-Consistency Multimodal Relevance Model for Search Ads](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BHCMRM%5D%20HCMRM%20-A%20High-Consistency%20Multimodal%20Relevance%20Model%20for%20Search%20Ads.pdf) \u003Cbr \u002F>\n* [2025 (Linkedin) (Arxiv) Powering Job Search at Scale - LLM-Enhanced Query Understanding in Job Matching Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%28Linkedin%29%20%28Arxiv%29%20Powering%20Job%20Search%20at%20Scale%20-%20LLM-Enhanced%20Query%20Understanding%20in%20Job%20Matching%20Systems.pdf) \u003Cbr \u002F>\n* [2025 （Alibaba) (Arxiv) TaoSR1 - The Thinking Model for E-commerce Relevance Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%EF%BC%88Alibaba%29%20%28Arxiv%29%20TaoSR1%20-%20The%20Thinking%20Model%20for%20E-commerce%20Relevance%20Search.pdf) \u003Cbr \u002F>\n* [2025 （Tencent) (KDD) [GenFR] Applying Large Language Model For Relevance Search In Tencent](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%EF%BC%88Tencent%29%20%28KDD%29%20%5BGenFR%5D%20Applying%20Large%20Language%20Model%20For%20Relevance%20Search%20In%20Tencent.pdf) \u003Cbr \u002F>\n\n## 06_LLM\n\n\n#### 01_LLM_Classical\n* [2013 (Google) (NIPS) [Word2vec] Distributed Representations of Words and Phrases and their Compositionality](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2013%20%28Google%29%20%28NIPS%29%20%5BWord2vec%5D%20Distributed%20Representations%20of%20Words%20and%20Phrases%20and%20their%20Compositionality.pdf) \u003Cbr \u002F>\n* [2014 (Google) (NIPS) [Seq2Seq] Sequence to Sequence Learning with Neural Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2014%20%28Google%29%20%28NIPS%29%20%5BSeq2Seq%5D%20Sequence%20to%20Sequence%20Learning%20with%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2017 (Google) (NIPS) [Transformer] Attention Is All You Need](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2017%20%28Google%29%20%28NIPS%29%20%5BTransformer%5D%20Attention%20Is%20All%20You%20Need.pdf) \u003Cbr \u002F>\n* [2017 (OpenAI) (NIPS) [RLHF] Deep Reinforcement Learning from Human Preferences](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2017%20%28OpenAI%29%20%28NIPS%29%20%5BRLHF%5D%20Deep%20Reinforcement%20Learning%20from%20Human%20Preferences.pdf) \u003Cbr \u002F>\n* [2018 (OpenAI) (Arxiv) [GPT-1] Improving Language Understanding by Generative Pre-Training](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2018%20%28OpenAI%29%20%28Arxiv%29%20%5BGPT-1%5D%20Improving%20Language%20Understanding%20by%20Generative%20Pre-Training.pdf) \u003Cbr \u002F>\n* [2019 (Google) (NAACL) [Bert] BERT - Pre-training of Deep Bidirectional Transformers for Language Understanding](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2019%20%28Google%29%20%28NAACL%29%20%5BBert%5D%20BERT%20-%20Pre-training%20of%20Deep%20Bidirectional%20Transformers%20for%20Language%20Understanding.pdf) \u003Cbr \u002F>\n* [2019 (OpenAI) (Arxiv) [GPT-2] Language Models are Unsupervised Multitask Learners](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2019%20%28OpenAI%29%20%28Arxiv%29%20%5BGPT-2%5D%20Language%20Models%20are%20Unsupervised%20Multitask%20Learners.pdf) \u003Cbr \u002F>\n* [2020 (Arxiv) Scaling Laws for Neural Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2020%20%28Arxiv%29%20Scaling%20Laws%20for%20Neural%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2020 (Meta) (NIPS) [RAG] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2020%20%28Meta%29%20%28NIPS%29%20%5BRAG%5D%20Retrieval-Augmented%20Generation%20for%20Knowledge-Intensive%20NLP%20Tasks.pdf) \u003Cbr \u002F>\n* [2020 (OpenAI) (Arxiv) [GPT-3] Language Models are Few-Shot Learners](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2020%20%28OpenAI%29%20%28Arxiv%29%20%5BGPT-3%5D%20Language%20Models%20are%20Few-Shot%20Learners.pdf) \u003Cbr \u002F>\n* [2021 (Microsoft) (Arxiv) [LoRA] LoRA - Low-Rank Adaptation of Large Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2021%20%28Microsoft%29%20%28Arxiv%29%20%5BLoRA%5D%20LoRA%20-%20Low-Rank%20Adaptation%20of%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2022 (Google) (Arxiv) [PaLM] PaLM - Scaling Language Modeling with Pathways](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28Arxiv%29%20%5BPaLM%5D%20PaLM%20-%20Scaling%20Language%20Modeling%20with%20Pathways.pdf) \u003Cbr \u002F>\n* [2022 (Google) (JMLR) [SwitchTransfomers] Switch Transformers - Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28JMLR%29%20%5BSwitchTransfomers%5D%20Switch%20Transformers%20-%20Scaling%20to%20Trillion%20Parameter%20Models%20with%20Simple%20and%20Efficient%20Sparsity.pdf) \u003Cbr \u002F>\n* [2022 (Google) (NIPS) [COT] Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28NIPS%29%20%5BCOT%5D%20Chain-of-Thought%20Prompting%20Elicits%20Reasoning%20in%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2022 (Google) (NIPS) [ChainOfThought] Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28NIPS%29%20%5BChainOfThought%5D%20Chain-of-Thought%20Prompting%20Elicits%20Reasoning%20in%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2022 (Google) (TMLR) [Emergent] Emergent Abilities of Large Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28TMLR%29%20%5BEmergent%5D%20Emergent%20Abilities%20of%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2022 (OpenAI) (Arxiv) [InstructGPT] [RLHF] Training language models to follow instructions with human feedback](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28OpenAI%29%20%28Arxiv%29%20%5BInstructGPT%5D%20%5BRLHF%5D%20Training%20language%20models%20to%20follow%20instructions%20with%20human%20feedback.pdf) \u003Cbr \u002F>\n* [2022 (OpenAI) (Arxiv) [PPO] Proximal Policy Optimization Algorithms](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28OpenAI%29%20%28Arxiv%29%20%5BPPO%5D%20Proximal%20Policy%20Optimization%20Algorithms.pdf) \u003Cbr \u002F>\n* [2022 (OpenAI) (Arxiv) [WebGPT] Learning to summarize from human feedback](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28OpenAI%29%20%28Arxiv%29%20%5BWebGPT%5D%20Learning%20to%20summarize%20from%20human%20feedback.pdf) \u003Cbr \u002F>\n* [2022 (OpenAI) (Arxiv) [WebGPT] WebGPT - Browser-assisted question-answering with human feedback](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28OpenAI%29%20%28Arxiv%29%20%5BWebGPT%5D%20WebGPT%20-%20Browser-assisted%20question-answering%20with%20human%20feedback.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (Arxiv) [QWEN] QWEN Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2023%20%28Alibaba%29%20%28Arxiv%29%20%5BQWEN%5D%20QWEN%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2023 (Meta) (Arxiv) [LLaMA-2] Llama 2 - Open Foundation and Fine-Tuned ChatModels](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2023%20%28Meta%29%20%28Arxiv%29%20%5BLLaMA-2%5D%20Llama%202%20-%20Open%20Foundation%20and%20Fine-Tuned%20ChatModels.pdf) \u003Cbr \u002F>\n* [2023 (Meta) (Arxiv) [LLaMA] LLaMA - Open and Efficient Foundation Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2023%20%28Meta%29%20%28Arxiv%29%20%5BLLaMA%5D%20LLaMA%20-%20Open%20and%20Efficient%20Foundation%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2023 (OpenAI) (Arxiv) [GPT4] GPT-4 Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2023%20%28OpenAI%29%20%28Arxiv%29%20%5BGPT4%5D%20GPT-4%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2024 (Alibaba) (Arxiv) [QWEN2] QWEN2 Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2024%20%28Alibaba%29%20%28Arxiv%29%20%5BQWEN2%5D%20QWEN2%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2024 (Arxiv) [TinyLlama] Arxiv TinyLlama - An Open-Source Small Language Model](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2024%20%28Arxiv%29%20%5BTinyLlama%5D%20Arxiv%20TinyLlama%20-%20An%20Open-Source%20Small%20Language%20Model.pdf) \u003Cbr \u002F>\n* [2024 (DeepSeek) (Arxiv) [GRPO] DeepSeekMath - Pushing the Limits of Mathematical Reasoning in Open Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2024%20%28DeepSeek%29%20%28Arxiv%29%20%5BGRPO%5D%20DeepSeekMath%20-%20Pushing%20the%20Limits%20of%20Mathematical%20Reasoning%20in%20Open%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [QWEN-2.5] QWEN 2.5 Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BQWEN-2.5%5D%20QWEN%202.5%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [Qwen2.5-VL] Qwen2.5-VL Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BQwen2.5-VL%5D%20Qwen2.5-VL%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [Qwen3 Embedding] Qwen3 Embedding - Advancing Text Embedding and Reranking Through Foundation Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BQwen3%20Embedding%5D%20Qwen3%20Embedding%20-%20Advancing%20Text%20Embedding%20and%20Reranking%20Through%20Foundation%20Models.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [Qwen3] Qwen3 Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BQwen3%5D%20Qwen3%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025 (Arxiv) A Survey of Large Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Arxiv%29%20A%20Survey%20of%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2025 (DeepSeek) (Nature) [DeepSeek-R1] DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28DeepSeek%29%20%28Nature%29%20%5BDeepSeek-R1%5D%20DeepSeek-R1%20incentivizes%20reasoning%20in%20LLMs%20through%20reinforcement%20learning.pdf) \u003Cbr \u002F>\n* [2025 (DeepSeek) [DeepSeek-R1] DeepSeek-R1 -Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28DeepSeek%29%20%5BDeepSeek-R1%5D%20DeepSeek-R1%20-Incentivizing%20Reasoning%20Capability%20in%20LLMs%20via%20Reinforcement%20Learning.pdf) \u003Cbr \u002F>\n* [2025 (DeepSeek) [DeepSeek-V3] DeepSeek-V3 Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28DeepSeek%29%20%5BDeepSeek-V3%5D%20DeepSeek-V3%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025 (Google) (Arxiv) [SigLIP2] SigLIP 2 - Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Google%29%20%28Arxiv%29%20%5BSigLIP2%5D%20SigLIP%202%20-%20Multilingual%20Vision-Language%20Encoders%20with%20Improved%20Semantic%20Understanding%2C%20Localization%2C%20and%20Dense%20Features.pdf) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FBook) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FMOE) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FModelOptimization) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FMultiModal) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FQuant) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FSFT) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FSpecificApplication) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FSurvey) \u003Cbr \u002F>\n* [resources](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002Fresources.txt) \u003Cbr \u002F>\n\n#### 02_Self_Supervised_Learning\n* [2020 (Alibaba) (AAAI) [DMR] Deep Match to Rank Model for Personalized Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Alibaba%29%20%28AAAI%29%20%5BDMR%5D%20Deep%20Match%20to%20Rank%20Model%20for%20Personalized%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (CIKM) [BERT4Rec] BERT4Rec - Sequential Recommendation with Bidirectional Encoder Representations from Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Alibaba%29%20%28CIKM%29%20%5BBERT4Rec%5D%20BERT4Rec%20-%20Sequential%20Recommendation%20with%20Bidirectional%20Encoder%20Representations%20from%20Transformer.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (KDD) Disentangled Self-Supervision in Sequential Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Alibaba%29%20%28KDD%29%20Disentangled%20Self-Supervision%20in%20Sequential%20Recommenders.pdf) \u003Cbr \u002F>\n* [2020 (Arxiv) UserBERT - Self-supervised User Representation Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Arxiv%29%20UserBERT%20-%20Self-supervised%20User%20Representation%20Learning.pdf) \u003Cbr \u002F>\n* [2020 (Arxiv) [SGL] Self-supervised Graph Learning for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Arxiv%29%20%5BSGL%5D%20Self-supervised%20Graph%20Learning%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2020 (CIKM) [S3Rec] S3-Rec - Self-Supervised Learning for Sequential Recommendation with Mutual Information Maximization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28CIKM%29%20%5BS3Rec%5D%20S3-Rec%20-%20Self-Supervised%20Learning%20for%20Sequential%20Recommendation%20with%20Mutual%20Information%20Maximization.pdf) \u003Cbr \u002F>\n* [2020 (EMNLP) [PTUM] PTUM - Pre-training User Model from Unlabeled User Behaviors via Self-supervision](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28EMNLP%29%20%5BPTUM%5D%20PTUM%20-%20Pre-training%20User%20Model%20from%20Unlabeled%20User%20Behaviors%20via%20Self-supervision.pdf) \u003Cbr \u002F>\n* [2020 (SIGIR) Self-Supervised Reinforcement Learning for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28SIGIR%29%20Self-Supervised%20Reinforcement%20Learning%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (Arxiv) [CLRec] Contrastive Learning for Debiased Candidate Generation in Large-Scale Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28Alibaba%29%20%28Arxiv%29%20%5BCLRec%5D%20Contrastive%20Learning%20for%20Debiased%20Candidate%20Generation%20in%20Large-Scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (CIKM) * [ZEUS] Self-Supervised Learning on Users’ Spontaneous Behaviors for Multi-Scenario Ranking in E-commerce](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%2A%20%5BZEUS%5D%20Self-Supervised%20Learning%20on%20Users%E2%80%99%20Spontaneous%20Behaviors%20for%20Multi-Scenario%20Ranking%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (WWW) Contrastive Pre-training for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28Alibaba%29%20%28WWW%29%20Contrastive%20Pre-training%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2021 (Google) (CIKM) Self-supervised Learning for Large-scale Item Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28Google%29%20%28CIKM%29%20Self-supervised%20Learning%20for%20Large-scale%20Item%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021 (WSDM) [Prop] PROP - Pre-training with Representative Words Prediction for Ad-hoc Retrieval](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28WSDM%29%20%5BProp%5D%20PROP%20-%20Pre-training%20with%20Representative%20Words%20Prediction%20for%20Ad-hoc%20Retrieval.pdf) \u003Cbr \u002F>\n\n## 07_Reinforcement_Learning\n* [2010 (Yahoo) (WWW) [LinUCB] A Contextual-Bandit Approach to Personalized News Article Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2010%20%28Yahoo%29%20%28WWW%29%20%5BLinUCB%5D%20A%20Contextual-Bandit%20Approach%20to%20Personalized%20News%20Article%20Recommendation.pdf) \u003Cbr \u002F>\n* [2018 (Spotify) (Recsys) [Spotify Bandit] Explore, Exploit, and Explain Personalizing Explainable Recommendations with Bandits](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2018%20%28Spotify%29%20%28Recsys%29%20%5BSpotify%20Bandit%5D%20Explore%2C%20Exploit%2C%20and%20Explain%20Personalizing%20Explainable%20Recommendations%20with%20Bandits.pdf) \u003Cbr \u002F>\n* [2018 [Microsoft] (WWW) [DRN] DRN - A Deep Reinforcement Learning Framework for News Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2018%20%5BMicrosoft%5D%20%28WWW%29%20%5BDRN%5D%20DRN%20-%20A%20Deep%20Reinforcement%20Learning%20Framework%20for%20News%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019 (Google) (IJCAI) *[SlateQ] SLATEQ - A Tractable Decomposition for Reinforcement Learning with Recommendation Sets](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2019%20%28Google%29%20%28IJCAI%29%20%2A%5BSlateQ%5D%20SLATEQ%20-%20A%20Tractable%20Decomposition%20for%20Reinforcement%20Learning%20with%20Recommendation%20Sets.pdf) \u003Cbr \u002F>\n* [2019 (Google) (WSDM) *[Top-K Off-Policy] Top-K Off-Policy Correction for a REINFORCE Recommender System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2019%20%28Google%29%20%28WSDM%29%20%2A%5BTop-K%20Off-Policy%5D%20Top-K%20Off-Policy%20Correction%20for%20a%20REINFORCE%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2019 (Sigweb) Deep Reinforcement Learning for Search, Recommendation, and Online Advertising - A Survey](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2019%20%28Sigweb%29%20Deep%20Reinforcement%20Learning%20for%20Search%2C%20Recommendation%2C%20and%20Online%20Advertising%20-%20A%20Survey.pdf) \u003Cbr \u002F>\n* [2020 (Bytedance) (KDD) [RAM] Jointly Learning to Recommend and Advertise](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2020%20%28Bytedance%29%20%28KDD%29%20%5BRAM%5D%20Jointly%20Learning%20to%20Recommend%20and%20Advertise.pdf) \u003Cbr \u002F>\n* [2020 (JD) (SIGIR) [NICF] Neural Interactive Collaborative Filtering](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2020%20%28JD%29%20%28SIGIR%29%20%5BNICF%5D%20Neural%20Interactive%20Collaborative%20Filtering.pdf) \u003Cbr \u002F>\n\n#### RL_classical\n* [1992 (ML) [REINFORCE] Simple statistical gradient-following algorithms for connectionist reinforcement learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F1992%20%28ML%29%20%5BREINFORCE%5D%20Simple%20statistical%20gradient-following%20algorithms%20for%20connectionist%20reinforcement%20learning.pdf) \u003Cbr \u002F>\n* [1999 (NIPS) [Actor-Critic] Actor-Critic Algorithms](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F1999%20%28NIPS%29%20%5BActor-Critic%5D%20Actor-Critic%20Algorithms.pdf) \u003Cbr \u002F>\n* [2013 (DeepMind) (Arxiv) [DQN] Playing Atari with Deep Reinforcement Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2013%20%28DeepMind%29%20%28Arxiv%29%20%5BDQN%5D%20Playing%20Atari%20with%20Deep%20Reinforcement%20Learning.pdf) \u003Cbr \u002F>\n* [2015 (DeepMind) (AAAI) [Double Q-learning] Deep Reinforcement Learning with Double Q-learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2015%20%28DeepMind%29%20%28AAAI%29%20%5BDouble%20Q-learning%5D%20Deep%20Reinforcement%20Learning%20with%20Double%20Q-learning.pdf) \u003Cbr \u002F>\n* [2015 (DeepMind) (Nature) [DQN] Human-level control through deep reinforcement learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2015%20%28DeepMind%29%20%28Nature%29%20%5BDQN%5D%20Human-level%20control%20through%20deep%20reinforcement%20learning.pdf) \u003Cbr \u002F>\n* [2016 (Google) (Arxiv) [A3C] Asynchronous Methods for Deep Reinforcement Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2016%20%28Google%29%20%28Arxiv%29%20%5BA3C%5D%20Asynchronous%20Methods%20for%20Deep%20Reinforcement%20Learning.pdf) \u003Cbr \u002F>\n* [2016 (OpenAI) (Nature) Mastering the game of Go with deep neural networks and tree search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2016%20%28OpenAI%29%20%28Nature%29%20Mastering%20the%20game%20of%20Go%20with%20deep%20neural%20networks%20and%20tree%20search.pdf) \u003Cbr \u002F>\n* [2017 (JMLR) [TRPO] Trust Region Policy Optimization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2017%20%28JMLR%29%20%5BTRPO%5D%20Trust%20Region%20Policy%20Optimization.pdf) \u003Cbr \u002F>\n* [2017 (OpenAI) (Arxiv) [PPO] Proximal Policy Optimization Algorithms](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2017%20%28OpenAI%29%20%28Arxiv%29%20%5BPPO%5D%20Proximal%20Policy%20Optimization%20Algorithms.pdf) \u003Cbr \u002F>\n* [2017 （OpenAI) (Arxiv) [ES] Evolution Strategies as a Scalable Alternative to Reinforcement Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2017%20%EF%BC%88OpenAI%29%20%28Arxiv%29%20%5BES%5D%20Evolution%20Strategies%20as%20a%20Scalable%20Alternative%20to%20Reinforcement%20Learning.pdf) \u003Cbr \u002F>\n\n## 08_Deep_Learning\n* [2012 (NIPS) [CNN] ImageNet Classification with Deep Convolutional Neural Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2012%20%28NIPS%29%20%5BCNN%5D%20ImageNet%20Classification%20with%20Deep%20Convolutional%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2014 (JMLR) [Dropout] Dropout - A Simple Way to Prevent Neural Networks from Overfitting](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2014%20%28JMLR%29%20%5BDropout%5D%20Dropout%20-%20A%20Simple%20Way%20to%20Prevent%20Neural%20Networks%20from%20Overfitting.pdf) \u003Cbr \u002F>\n* [2015 (Google) (JMLR) [BatchNorm] Batch Normalization - Accelerating Deep Network Training by Reducing Internal Covariate Shift](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2015%20%28Google%29%20%28JMLR%29%20%5BBatchNorm%5D%20Batch%20Normalization%20-%20Accelerating%20Deep%20Network%20Training%20by%20Reducing%20Internal%20Covariate%20Shift.pdf) \u003Cbr \u002F>\n* [2015 (OpenAI) (ICLR) [Adam] Adam - A Method for Stochastic Optimization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2015%20%28OpenAI%29%20%28ICLR%29%20%5BAdam%5D%20Adam%20-%20A%20Method%20for%20Stochastic%20Optimization.pdf) \u003Cbr \u002F>\n* [2016 (CVPR) [ResNet] Deep Residual Learning for Image Recognition](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2016%20%28CVPR%29%20%5BResNet%5D%20Deep%20Residual%20Learning%20for%20Image%20Recognition.pdf) \u003Cbr \u002F>\n* [2016 (OpenAI) (NIPS) [Weight Norm] Weight Normalization - A Simple Reparameterization to Accelerate Training of Deep Neural Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2016%20%28OpenAI%29%20%28NIPS%29%20%5BWeight%20Norm%5D%20Weight%20Normalization%20-%20A%20Simple%20Reparameterization%20to%20Accelerate%20Training%20of%20Deep%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2017 (Arxiv) [LayerNorm] Layer Normalization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2017%20%28Arxiv%29%20%5BLayerNorm%5D%20Layer%20Normalization.pdf) \u003Cbr \u002F>\n* [2017 (Google) (NIPS) [Transformer] Attention Is All You Need](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2017%20%28Google%29%20%28NIPS%29%20%5BTransformer%5D%20Attention%20Is%20All%20You%20Need.pdf) \u003Cbr \u002F>\n* [2020 (Arxiv) GLU Variants Improve Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2020%20%28Arxiv%29%20GLU%20Variants%20Improve%20Transformer.pdf) \u003Cbr \u002F>\n* [2020 (ICML) On Layer Normalization in the Transformer Architecture](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2020%20%28ICML%29%20On%20Layer%20Normalization%20in%20the%20Transformer%20Architecture.pdf) \u003Cbr \u002F>\n","## 面向工业界搜索、推荐和广告的优秀深度学习论文。这些论文主要关注嵌入、匹配、预排序、排序（CTR\u002FCVR预测）、后排序、相关性、大语言模型、强化学习等领域。\n\n## 00_嵌入\n* [2013年（谷歌）（NIPS）[Word2vec] 单词与短语的分布式表示及其组合性](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2013%20%28Google%29%20%28NIPS%29%20%5BWord2vec%5D%20Distributed%20Representations%20of%20Words%20and%20Phrases%20and%20their%20Compositionality.pdf) \u003Cbr \u002F>\n* [2014年（KDD）[DeepWalk] DeepWalk - 社交网络表示的在线学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2014%20%28KDD%29%20%5BDeepWalk%5D%20%20DeepWalk%20-%20online%20learning%20of%20social%20representations.pdf) \u003Cbr \u002F>\n* [2015年（WWW）[LINE] LINE 大规模信息网络嵌入](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2015%20%28WWW%29%20%5BLINE%5D%20LINE%20Large-scale%20Information%20Network%20Embedding.pdf) \u003Cbr \u002F>\n* [2016年（KDD）[Node2vec] node2vec - 可扩展的网络特征学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2016%20%28KDD%29%20%5BNode2vec%5D%20node2vec%20-%20Scalable%20Feature%20Learning%20for%20Networks.pdf) \u003Cbr \u002F>\n* [2017年（ICLR）[GCN] 图卷积网络的半监督分类](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2017%20%28ICLR%29%20%5BGCN%5D%20Semi-supervised%20Classification%20with%20Graph%20Convolutional%20Networks%20.pdf) \u003Cbr \u002F>\n* [2017年（KDD）[Struc2vec] struc2vec - 从结构身份中学习节点表示](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2017%20%28KDD%29%20%5BStruc2vec%5D%20struc2vec%20-%20Learning%20Node%20Representations%20from%20Structural%20Identity.pdf) \u003Cbr \u002F>\n* [2017年（NIPS）[GraphSAGE] 大规模图上的归纳式表示学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2017%20%28NIPS%29%20%5BGraphSAGE%5D%20Inductive%20Representation%20Learning%20on%20Large%20Graphs.pdf) \u003Cbr \u002F>\n* [2018年（阿里巴巴）（KDD）*[阿里巴巴嵌入] 阿里巴巴电商推荐中的百亿级商品嵌入](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2018%20%28Alibaba%29%20%28KDD%29%20%2A%5BAlibaba%20Embedding%5D%20Billion-scale%20Commodity%20Embedding%20for%20E-commerce%20Recommendation%20in%20Alibaba.pdf) \u003Cbr \u002F>\n* [2018年（ICLR）[GAT] 图注意力网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2018%20%28ICLR%29%20%5BGAT%5D%20%20Graph%20Attention%20Networks.pdf) \u003Cbr \u002F>\n* [2018年（Pinterest）（KDD）*[PinSage] 用于大规模推荐系统的图卷积神经网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2018%20%28Pinterest%29%20%28KDD%29%20%2A%5BPinSage%5D%20Graph%20Convolutional%20Neural%20Networks%20for%20Web-Scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2018年（WSDM）[NetMF] 网络嵌入即矩阵分解 - 统一deepwalk、line、pte和node2vec](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2018%20%28WSDM%29%20%5BNetMF%5D%20Network%20embedding%20as%20matrix%20factorization%20-%20Unifying%20deepwalk%2C%20line%2C%20pte%2C%20and%20node2vec.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（KDD）*[GATNE] 带属性的多层异构网络表示学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F00_Embedding\u002F2019%20%28Alibaba%29%20%28KDD%29%20%2A%5BGATNE%5D%20Representation%20Learning%20for%20Attributed%20Multiplex%20Heterogeneous%20Network.pdf) \u003Cbr \u002F>\n\n## 01_匹配\n* [1994年（CSCW）[用户协同过滤] GroupLens - 面向Netnews协同过滤的开放架构](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F1994%20%28CSCW%29%20%5BUser-CF%5D%20GroupLens%20-%20An%20Open%20Architecture%20for%20Collaborative%20Filtering%20of%20Netnews.pdf) \u003Cbr \u002F>\n* [1998年（微软）协同过滤预测算法的实证分析](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F1998%20%28Microsoft%29%20Empirical%20Analysis%20of%20Predictive%20Algorithms%20for%20Collaborative%20Filtering.pdf) \u003Cbr \u002F>\n* [2003年（亚马逊）[物品协同过滤] Amazon.com推荐——基于物品的协同过滤](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2003%20%28Amazon%29%20%5BItem-CF%5D%20Amazon.com%20recommendations%20-%20item-to-item%20collaborative%20filtering.pdf) \u003Cbr \u002F>\n* [2009年（Computer）[矩阵分解] 推荐系统中的矩阵分解技术](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2009%20%28Computer%29%20%5BMF%5D%20Matrix%20factorization%20techniques%20for%20recommender%20systems.pdf) \u003Cbr \u002F>\n* [2013年（微软）（CIKM）[DSSM] 使用点击数据学习用于网络搜索的深度结构化语义模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2013%20%28Microsoft%29%20%28CIKM%29%20%5BDSSM%5D%20Learning%20deep%20structured%20semantic%20models%20for%20web%20search%20using%20clickthrough%20data.pdf) \u003Cbr \u002F>\n* [2015年（KDD）[Sceptre] 推断可替代与互补产品的网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2015%20%28KDD%29%20%5BSceptre%5D%20Inferring%20Networks%20of%20Substitutable%20and%20Complementary%20Products.pdf) \u003Cbr \u002F>\n* [2016年（谷歌）（RecSys）**[YouTube DNN] 用于YouTube推荐的深度神经网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2016%20%28Google%29%20%28RecSys%29%20%2A%2A%5BYoutube%20DNN%5D%20Deep%20Neural%20Networks%20for%20YouTube%20Recommendations.pdf) \u003Cbr \u002F>\n* [2018年（Airbnb）（KDD）*[Airbnb Embedding] 使用嵌入进行实时个性化：Airbnb搜索排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2018%20%28Airbnb%29%20%28KDD%29%20%2A%5BAirbnb%20Embedding%5D%20Real-time%20Personalization%20using%20Embeddings%20for%20Search%20Ranking%20at%20Airbnb.pdf) \u003Cbr \u002F>\n* [2018年（阿里巴巴）（KDD）* [TDM] 学习用于推荐系统的树形深度模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2018%20%28Alibaba%29%20%28KDD%29%20%2A%20%5BTDM%5D%20Learning%20Tree-based%20Deep%20Model%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2018年（Pinterest）（KDD）*[PinSage] 用于大规模推荐系统的图卷积神经网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2018%20%28Pinterest%29%20%28KDD%29%20%2A%5BPinSage%5D%20Graph%20Convolutional%20Neural%20Networks%20for%20Web-Scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（CIKM）**[MIND] 基于动态路由的多兴趣网络，用于天猫推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Alibaba%29%20%28CIKM%29%20%2A%2A%5BMIND%5D%20Multi-Interest%20Network%20with%20Dynamic%20Routing%20for%20Recommendation%20at%20Tmall.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（CIKM）*[SDM] SDM——面向在线大规模推荐系统的序列式深度匹配模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Alibaba%29%20%28CIKM%29%20%2A%5BSDM%5D%20SDM%20-%20Sequential%20deep%20matching%20model%20for%20online%20large-scale%20recommender%20system.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（NIPS）*[JTM] 推荐系统中基于树的索引与深度模型的联合优化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Alibaba%29%20%28NIPS%29%20%2A%5BJTM%5D%20Joint%20Optimization%20of%20Tree-based%20Index%20and%20Deep%20Model%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2019年（亚马逊）（KDD）语义产品搜索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Amazon%29%20%28KDD%29%20Semantic%20Product%20Search.pdf) \u003Cbr \u002F>\n* [2019年（百度）（KDD）*[MOBIUS] MOBIUS——迈向百度竞价排名中下一代查询与广告匹配](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Baidu%29%20%28KDD%29%20%2A%5BMOBIUS%5D%20MOBIUS%20-%20Towards%20the%20Next%20Generation%20of%20Query-Ad%20Matching%20in%20Baidu%27s%20Sponsored%20Search.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（RecSys）**[双塔] 针对大型语料库物品推荐的采样偏差校正神经网络模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Google%29%20%28RecSys%29%20%2A%2A%5BTwo-Tower%5D%20Sampling-Bias-Corrected%20Neural%20Modeling%20for%20Large%20Corpus%20Item%20Recommendations.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（WSDM）*[Top-K 离策略] 针对REINFORCE推荐系统的Top-K离策略校正](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%28Google%29%20%28WSDM%29%20%2A%5BTop-K%20Off-Policy%5D%20Top-K%20Off-Policy%20Correction%20for%20a%20REINFORCE%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2019年（腾讯）（KDD）腾讯查询与文档理解中的以用户为中心的概念挖掘系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2019%20%5BTencent%5D%20%28KDD%29%20A%20User-Centered%20Concept%20Mining%20System%20for%20Query%20and%20Document%20Understanding%20at%20Tencent.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（Arxiv）[SWING] 用于电子商务推荐的大规模商品图构建](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Alibaba%29%20%28Arxiv%29%20%5BSWING%5D%20Large%20Scale%20Product%20Graph%20Construction%20for%20Recommendation%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（ICML）[OTM] 在束搜索下学习最优树模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Alibaba%29%20%28ICML%29%20%5BOTM%5D%20Learning%20Optimal%20Tree%20Models%20under%20Beam%20Search.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（KDD）*[ComiRec] 可控多兴趣框架用于推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Alibaba%29%20%28KDD%29%20%2A%5BComiRec%5D%20Controllable%20Multi-Interest%20Framework%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2020年（Facebook）（KDD）**[Facebook搜索中的嵌入] Facebook搜索中的基于嵌入的召回](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Facebook%29%20%28KDD%29%20%2A%2A%5BEmbedding%20for%20Facebook%20Search%5D%20Embedding-based%20Retrieval%20in%20Facebook%20Search.pdf) \u003Cbr \u002F>\n* [2020年（谷歌）（WWW）*[MNS] 混合负采样用于学习推荐中的双塔神经网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Google%29%20%28WWW%29%20%2A%5BMNS%5D%20Mixed%20Negative%20Sampling%20for%20Learning%20Two-tower%20Neural%20Networks%20in%20Recommendations.pdf) \u003Cbr \u002F>\n* [2020年（京东）（CIKM）*[DecGCN] 用于推断可替代与互补商品的解耦图卷积网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28JD%29%20%28CIKM%29%20%2A%5BDecGCN%5D%20Decoupled%20Graph%20Convolution%20Network%20for%20Inferring%20Substitutable%20and%20Complementary%20Items.pdf) \u003Cbr \u002F>\n* [2020年（京东）（SIGIR）[DPSR] 向个性化与语义召回迈进——通过嵌入学习实现电商搜索的端到端解决方案](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28JD%29%20%28SIGIR%29%20%2A%5BDPSR%5D%20Towards%20Personalized%20and%20Semantic%20Retrieval%20-%20An%20End-to-EndSolution%20for%20E-commerce%20Search%20via%20Embedding%20Learning.pdf) \u003Cbr \u002F>\n* [2020年（微软）（Arxiv）TwinBERT——将知识蒸馏至孪生结构的BERT模型，以实现高效检索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2020%20%28Microsoft%29%20%28Arxiv%29%20TwinBERT%20-%20Distilling%20Knowledge%20to%20Twin-Structured%20BERT%20Models%20for%20Efficient%20Retrieval.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（KDD）* [MGDSPR] 淘宝搜索中的基于嵌入的商品召回](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Alibaba%29%20%28KDD%29%20%2A%20%5BMGDSPR%5D%20Embedding-based%20Product%20Retrieval%20in%20Taobao%20Search.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（SIGIR）* [PDN] 基于路径的深度网络，用于推荐中的候选商品匹配](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Alibaba%29%20%28SIGIR%29%20%2A%20%5BPDN%5D%20Path-based%20Deep%20Network%20for%20Candidate%20Item%20Matching%20in%20Recommenders.pdf) \u003Cbr \u002F>\n* [2021年（亚马逊）（KDD）产品搜索中语义匹配的极端多标签学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Amazon%29%20%28KDD%29%20Extreme%20Multi-label%20Learning%20for%20Semantic%20Matching%20in%20Product%20Search.pdf) \u003Cbr \u002F>\n* [2021年（百度）（KDD）百度搜索中面向网络规模召回的预训练语言模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Baidu%29%20%28KDD%29%20Pre-trained%20Language%20Model%20for%20Web-scale%20Retrieval%20in%20Baidu%20Search.pdf) \u003Cbr \u002F>\n* [2021年（字节跳动）（Arxiv）[DR] 深度召回——为大规模推荐学习可检索结构](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Bytedance%29%20%28Arxiv%29%20%5BDR%5D%20Deep%20Retrieval%20-%20Learning%20A%20Retrievable%20Structure%20for%20Large-Scale%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021年（美团）（DLP-KDD）[DAT] 用于在线大规模推荐的双重增强双塔模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2021%20%28Meituan%29%20%28DLP-KDD%29%20%5BDAT%5D%20A%20Dual%20Augmented%20Two-tower%20Model%20for%20Online%20Large-scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（CIKM）**[NANN] 大规模推荐中基于神经相似性度量的近似最近邻搜索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2022%20%28Alibaba%29%20%28CIKM%29%20%2A%2A%5BNANN%5D%20Approximate%20Nearest%20Neighbor%20Search%20under%20Neural%20Similarity%20Metric%20for%20Large-Scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（CIKM）[CLE-QR] 淘宝搜索中的查询改写](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2022%20%28Alibaba%29%20%28CIKM%29%20%5BCLE-QR%5D%20Query%20Rewriting%20in%20TaoBao%20Search.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）**（CIKM）[MOPPR] 淘宝搜索中的多目标个性化商品召回](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2022%20%28Alibaba%29%20%2A%2A%28CIKM%29%20%5BMOPPR%5D%20Multi-Objective%20Personalized%20Product%20Retrieval%20in%20Taobao%20Search.pdf) \u003Cbr \u002F>\n* [2024年（字节跳动）（KDD）[Trinity] Trinity——将多兴趣、长尾及长期兴趣融为一体](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2024%20%28Bytedance%29%20%28KDD%29%20%5BTrinity%5D%20Trinity%20-%20Syncretizing%20Multi-%3ALong-Tail%3ALong-Term%20Interests%20All%20in%20One.pdf) \u003Cbr \u002F>\n* [2024年（Meta）（Arxiv）** [GR] 行动胜于言语——用于生成式推荐的万亿参数序列转换器](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2024%20%28Meta%29%20%28Arxiv%29%20%2A%2A%20%5BGR%5D%20Actions%20Speak%20Louder%20than%20Words%20-%20Trillion-Parameter%20Sequential%20Transducers%20for%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）（Arxiv）[LongRetriever] LongRetriever——迈向基于超长序列的推荐候选召回](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Bytedance%29%20%28Arxiv%29%20%5BLongRetriever%5D%20LongRetriever%20-%20Towards%20Ultra-Long%20Sequence%20based%20Candidate%20Retrieval%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）（KDD）[VQ] 基于流式向量量化检索器的实时索引，用于大规模推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Bytedance%29%20%28KDD%29%20%5BVQ%5D%20Real-time%20Indexing%20for%20Large-scale%20Recommendation%20by%20Streaming%20Vector%20Quantization%20Retriever.pdf) \u003Cbr \u002F>\n* [2025年（京东）（KDD）[UniERF] UniERF——面向电商搜索的统一嵌入式召回框架](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28JD%29%20%28KDD%29%20%5BUniERF%5D%20UniERF%20-%20A%20Uniform%20Embedding-based%20Retrieval%20Framework%20for%20E-Commerce%20Search.pdf) \u003Cbr \u002F>\n* [2025年（Meta）（KDD）[MTMH] 提升召回还是相关性？推荐中基于多任务多头的物品间召回方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Meta%29%20%28KDD%29%20%5BMTMH%5D%20Optimizing%20Recall%20or%20Relevance%3F%20A%20Multi-Task%20Multi-Head%20Approach%20for%20Item-to-Item%20Retrieval%20in%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025年（Meta）[RADAR ] RADAR——通过延迟异步召回提升召回率](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Meta%29%20%5BRADAR%20%5D%20RADAR%20-%20Recall%20Augmentation%20through%20Deferred%20Asynchronous%20Retrieval.pdf) \u003Cbr \u002F>\n* [2025年（腾讯）（Arxiv）一种高效的GPU加速特征交互的嵌入式广告召回](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002F2025%20%28Tencent%29%20%28Arxiv%29%20An%20Efficient%20Embedding%20Based%20Ad%20Retrieval%20with%20GPU-Powered%20Feature%20Interaction.pdf) \u003Cbr \u002F>\n\n#### 近似最近邻搜索\n* [2017 (Arxiv) (Meta) [FAISS] 基于GPU的十亿级相似度搜索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FANN\u002F2017%20%28Arxiv%29%20%28Meta%29%20%5BFAISS%5D%20Billion-scale%20similarity%20search%20with%20GPUs.pdf) \u003Cbr \u002F>\n* [2020 (PAMI) [HNSW] 使用分层可导航小世界图的高效且鲁棒的近似最近邻搜索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FANN\u002F2020%20%28PAMI%29%20%5BHNSW%5D%20Efficient%20and%20Robust%20Approximate%20Nearest%20Neighbor%20Search%20Using%20Hierarchical%20Navigable%20Small%20World%20Graphs.pdf) \u003Cbr \u002F>\n* [2021 (TPAMI) [IVF-PQ] 用于最近邻搜索的产品量化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FANN\u002F2021%20%28TPAMI%29%20%5BIVF-PQ%5D%20Product%20Quantization%20for%20Nearest%20Neighbor%20Search.pdf) \u003Cbr \u002F>\n\n#### 图神经网络\n* [2017 (ICLR) [GCN] 基于图卷积网络的半监督分类](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2017%20%28ICLR%29%20%5BGCN%5D%20Semi-Supervised%20Classification%20with%20Graph%20Convolutional%20Networks.pdf) \u003Cbr \u002F>\n* [2018 (ICLR) [GAT] 图注意力网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2018%20%28ICLR%29%20%5BGAT%5D%20Graph%20Attention%20Networks.pdf) \u003Cbr \u002F>\n* [2018 (Pinterest) (KDD) [PinSage] 面向大规模推荐系统的图卷积神经网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2018%20%28Pinterest%29%20%28KDD%29%20%5BPinSage%5D%20Graph%20Convolutional%20Neural%20Networks%20for%20Web-Scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) [IntentGC] IntentGC——一种融合异构信息的可扩展图卷积框架，用于推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BIntentGC%5D%20IntentGC%20-%20a%20Scalable%20Graph%20Convolution%20Framework%20Fusing%20Heterogeneous%20Information%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (KDD) [MEIRec] 基于元路径引导的异构图神经网络，用于意图推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BMEIRec%5D%20Metapath-guided%20Heterogeneous%20Graph%20Neural%20Network%20for%20Intent%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (SIGIR) [GIN] 用于赞助搜索中点击率预测的图意图网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2019%20%28Alibaba%29%20%28SIGIR%29%20%5BGIN%5D%20Graph%20Intention%20Network%20for%20Click-through%20Rate%20Prediction%20in%20Sponsored%20Search.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (SIGIR) [ATBRG] ATBRG——一种自适应目标-行为关系图网络，用于有效推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FGraph_Neural_Networks\u002F2020%20%28Alibaba%29%20%28SIGIR%29%20%5BATBRG%5D%20ATBRG%20-%20Adaptive%20Target-Behavior%20Relational%20Graph%20Network%20for%20Effective%20Recommendation.pdf) \u003Cbr \u002F>\n\n#### LLM_匹配\n* [2021年（百度）（KDD）用于百度搜索中Web规模检索的预训练语言模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2021%20%28Baidu%29%20%28KDD%29%20Pre-trained%20Language%20Model%20for%20Web-scale%20Retrieval%20in%20Baidu%20Search.pdf) \u003Cbr \u002F>\n* [2023年（谷歌）（NIPS）[TIGER] 基于生成式检索的推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2023%20%28Google%29%20%28NIPS%29%20%5BTIGER%5D%20Recommender%20Systems%20with%20Generative%20Retrieval.pdf) \u003Cbr \u002F>\n* [2024年（阿里巴巴）（WWW）[BEQUE] 淘宝搜索中基于大语言模型的长尾查询改写](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Alibaba%29%20%28WWW%29%20%5BBEQUE%5D%20Large%20Language%20Model%20based%20Long-tail%20Query%20Rewriting%20in%20Taobao%20Search.pdf) \u003Cbr \u002F>\n* [2024年（快手）（Arxiv）[KuaiFormer] KuaiFormer - 快手基于Transformer的检索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Kuaishou%29%20%28Arxiv%29%20%5BKuaiFormer%5D%20KuaiFormer%20-%20Transformer-Based%20Retrieval%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2024年（Meta）（Arxiv）** [GR] 行动胜于言辞 - 用于生成式推荐的万亿参数序列转换器](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Meta%29%20%28Arxiv%29%20%2A%2A%20%5BGR%5D%20Actions%20Speak%20Louder%20than%20Words%20-%20Trillion-Parameter%20Sequential%20Transducers%20for%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024年（Meta）（Arxiv）统一生成式与稠密检索用于序列推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Meta%29%20%28Arxiv%29%20Unifying%20Generative%20and%20Dense%20Retrieval%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024年（小红书）（WWW）[NoteLLM] NoteLLM - 用于推荐的多模态大型表示模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2024%20%28Xiaohongshu%29%20%28WWW%29%20%5BNoteLLM%5D%20NoteLLM%20-%20Multimodal%20Large%20Representation%20Models%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025年（快手）（Arxiv）[OneRec] OneRec - 统一检索与排序，结合生成式推荐与偏好对齐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneRec%5D%20OneRec%20-%20Unifying%20Retrieve%20and%20Rank%20with%20Generative%20Recommender%20and%20Preference%20Alignment.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（Arxiv）[TBGRecall] TBGRecall - 面向电商推荐场景的生成式检索模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BTBGRecall%5D%20TBGRecall%20-%20A%20Generative%20Retrieval%20Model%20for%20E-commerce%20Recommendation%20Scenarios.pdf) \u003Cbr \u002F>\n* [2025年（腾讯）（Arxiv）[RARE] 基于大语言模型生成商业意图的实时广告检索，用于赞助搜索广告](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Arxiv%29%20%28Tencent%29%20%5BRARE%5D%20Real-time%20Ad%20retrieval%20via%20LLM-generative%20Commercial%20Intention%20for%20Sponsored%20Search%20Advertising.pdf) \u003Cbr \u002F>\n* [2025年（百度）（Arxiv）[COBRA] 稀疏遇见稠密 - 结合级联稀疏-稠密表征的统一生成式推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Baidu%29%20%28Arxiv%29%20%5BCOBRA%5D%20Sparse%20Meets%20Dense%20-Unified%20Generative%20Recommendations%20with%20Cascaded%20Sparse-Dense%20Representations.pdf) \u003Cbr \u002F>\n* [2025年（谷歌）[PLUM] PLUM - 为工业规模生成式推荐适配预训练语言模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Google%29%20%5BPLUM%5D%20PLUM%20-%20Adapting%20Pre-trained%20Language%20Models%20for%20Industrial-scale%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2025年（京东）（Arxiv）[GRAM] 生成式检索与对齐模型 - 电商检索的新范式](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28JD%29%20%28Arxiv%29%20%5BGRAM%5D%20Generative%20Retrieval%20and%20Alignment%20Model%20-%20A%20New%20Paradigm%20for%20E-commerce%20Retrieval.pdf) \u003Cbr \u002F>\n* [2025年（快手）（AAAI）[Align3GR] Align3GR - 面向基于大语言模型的生成式推荐的统一多层级对齐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Kuaishou%29%20%28AAAI%29%20%5BAlign3GR%5D%20Align3GR%20-%20Unified%20Multi-Level%20Alignment%20for%20LLM-based%20Generative%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025年（快手）（Arxiv）[LARM] LLM对齐直播推荐pdf](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BLARM%5D%20LLM-Alignment%20Live-Streaming%20Recommendationpdf.pdf) \u003Cbr \u002F>\n* [2025年（快手）（Arxiv）[LEARN] LEARN - 将大语言模型的知识适配到推荐系统中，以实现实际工业应用](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BLEARN%5D%20LEARN%20-%20Knowledge%20Adaptation%20from%20Large%20Language%20Model%20to%20Recommendation%20for%20Practical%20Industrial%20Application.pdf) \u003Cbr \u002F>\n* [2025年（Meta）（Arxiv）[DRAMA] DRAMA - 从大语言模型到小型稠密检索器的多样化增强](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Meta%29%20%28Arxiv%29%20%5BDRAMA%5D%20DRAMA%20-%20Diverse%20Augmentation%20from%20Large%20Language%20Models%20to%20Smaller%20Dense%20Retrievers.pdf) \u003Cbr \u002F>\n* [2025年（Meta）（Arxiv）[ROO] 仅针对请求的推荐系统优化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Meta%29%20%28Arxiv%29%20%5BROO%5D%20Request-Only%20Optimization%20for%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025年（Pinterest）[PinRec] PinRec - 面向工业规模推荐系统的、基于结果条件的多令牌生成式检索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Pinterest%29%20%5BPinRec%5D%20PinRec%20-%20Outcome-Conditioned%2C%20Multi-Token%20Generative%20Retrieval%20for%20Industry-Scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025年（小红书）（KDD）[NoteLLM-2] NoteLLM-2 - 用于推荐的多模态大型表示模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F01_Matching\u002FLLM_Matching\u002F2025%20%28Xiaohongshu%29%20%EF%BC%88KDD%29%20%5BNoteLLM-2%5D%20NoteLLM-2%20-%20Multimodal%20Large%20Representation%20Models%20for%20Recommendation.pdf) \u003Cbr \u002F>\n\n## 02_预排序\n* [2020年（阿里巴巴）（DLP-KDD）[COLD] COLD - 迈向下一代预排序系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2020%20%28Alibaba%29%20%28DLP-KDD%29%20%5BCOLD%5D%20COLD%20-%20Towards%20the%20Next%20Generation%20of%20Pre-Ranking%20System.pdf) \u003Cbr \u002F>\n* [2022年（华为）（SIGIR）[RankFlow] RankFlow - 将多阶段级联排序系统作为流进行联合优化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2022%20%28Huawei%29%20%28SIGIR%29%20%5BRankFlow%5D%20RankFlow%20-%20JointOptimization%20ofMulti-Stage%20CascadeRanking%20SystemsasFlows.pdf) \u003Cbr \u002F>\n* [2022年（华为）（CIKM）[IntTower] IntTower - 预排序系统的下一代双塔模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2022%20%EF%BC%88Huawei%29%20%28CIKM%29%20%5BIntTower%5D%20IntTower%20-%20the%20Next%20Generation%20of%20Two-Tower%20Model%20for%20Pre-Ranking%20System.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（CIKM）[COPR] COPR - 面向一致性的在线广告预排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BCOPR%5D%20COPR%20-%20Consistency-Oriented%20Pre-Ranking%20for%20Online%20Advertising.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（KDD）[ASMOL] 重新思考大规模电商搜索系统中预排序的作用](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2023%20%28Alibaba%29%20%28KDD%29%20%5BASMOL%5D%20Rethinking%20the%20Role%20of%20Pre-ranking%20in%20Large-scale%20E-Commerce%20Searching%20System.pdf) \u003Cbr \u002F>\n* [2025年（腾讯）（Arxiv）[HIT] HIT模型 - 一种用于预排序系统的分层交互增强型双塔模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F02_Pre-ranking\u002F2025%20%28Tencent%29%20%28Arxiv%29%20%5BHIT%5D%20HIT%20Model%20-%20A%20Hierarchical%20Interaction-Enhanced%20Two-Tower%20Model%20for%20Pre-Ranking%20Systems.pdf) \u003Cbr \u002F>\n\n## 03_排序\n* [2014年（ADKDD）（Facebook）从Facebook广告点击率预测中获得的实践经验](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2014%20%28ADKDD%29%20%28Facebook%29%20Practical%20Lessons%20from%20Predicting%20Clicks%20on%20Ads%20at%20Facebook.pdf) \u003Cbr \u002F>\n* [2016年（Google）（DLRS）**[Wide & Deep] 面向推荐系统的宽深学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2016%20%28Google%29%20%28DLRS%29%20%2A%2A%5BWide%20%26%20Deep%5D%20Wide%20%26%20Deep%20Learning%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2016年（Google）（RecSys）**[Youtube DNN] 面向YouTube推荐的深度神经网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2016%20%28Google%29%20%28RecSys%29%20%2A%2A%5BYoutube%20DNN%5D%20Deep%20Neural%20Networks%20for%20YouTube%20Recommendations.pdf) \u003Cbr \u002F>\n* [2018年（阿里巴巴）（KDD）**[DIN] 用于点击率预测的深度兴趣网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2018%20%28Alibaba%29%20%28KDD%29%20%2A%2A%5BDIN%5D%20Deep%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（AAAI）**[DIEN] 用于点击率预测的深度兴趣演化网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2019%20%28Alibaba%29%20%28AAAI%29%20%2A%2A%5BDIEN%5D%20Deep%20Interest%20Evolution%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（CIKM）** [AutoInt] AutoInt - 基于自注意力神经网络的自动特征交互学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2019%20%28CIKM%29%20%2A%2A%20%5BAutoInt%5D%20AutoInt%20-Automatic%20Feature%20Interaction%20Learning%20via%20Self-Attentive%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2019年（Facebook）（Arxiv）[DLRM] （Facebook）面向个性化和推荐系统的深度学习推荐模型，Facebook](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2019%20%28Facebook%29%20%28Arxiv%29%20%5BDLRM%5D%20%28Facebook%29%20Deep%20Learning%20Recommendation%20Model%20for%20Personalization%20and%20Recommendation%20Systems%2C%20Facebook.pdf) \u003Cbr \u002F>\n* [2019年（Google）（Recsys）** [Youtube多任务] 推荐下一个要观看的视频——一个多任务排名系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2019%20%28Google%29%20%28Recsys%29%20%2A%2A%20%5BYoutube%20Multi-task%5D%20Recommending%20what%20video%20to%20watch%20next%20-%20a%20multitask%20ranking%20system.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（Arxiv）** [SIM] 基于搜索的用户兴趣建模：利用终身序列行为数据进行点击率预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2020%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BSIM%5D%20Search-based%20User%20Interest%20Modeling%20with%20Lifelong%20Sequential%20Behavior%20Data%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（NIPS）使用极化正则化的神经元级结构化剪枝](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2020%20%28Alibaba%29%20%28NIPS%29%20Neuron-level%20Structured%20Pruning%20using%20Polarization%20Regularizer.pdf) \u003Cbr \u002F>\n* [2020年（京东）（CIKM）**[DMT] 面向大规模电商推荐系统的多目标排序的深度多面Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2020%20%28JD%29%20%28CIKM%29%20%2A%2A%5BDMT%5D%20Deep%20Multifaceted%20Transformers%20for%20Multi-objective%20Ranking%20in%20Large-Scale%20E-commerce%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2020年（腾讯）（Recsys）** [PLE] 渐进式分层提取（PLE）——一种用于个性化推荐的新型多任务学习（MTL）模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2020%20%28Tencent%29%20%28Recsys%29%20%2A%2A%20%20%5BPLE%5D%20Progressive%20Layered%20Extraction%20%28PLE%29%20-%20A%20Novel%20Multi-Task%20Learning%20%28MTL%29%20Model%20for%20Personalized%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（CIKM）* [ZEUS] 面向电商多场景排序的用户自发行为自监督学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%2A%20%5BZEUS%5D%20Self-Supervised%20Learning%20on%20Users%E2%80%99%20Spontaneous%20Behaviors%20for%20Multi-Scenario%20Ranking%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（CIKM）[STAR] 一模型服务所有——面向多领域点击率预测的星型拓扑自适应推荐器](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%5BSTAR%5D%20One%20Model%20to%20Serve%20All%20-%20Star%20Topology%20Adaptive%20Recommender%20for%20Multi-Domain%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2021年（谷歌）（WWW）* [DCN V2] DCN V2 — 改进的深度交叉网络及面向Web规模排序学习系统的实践启示](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2021%20%28Google%29%20%28WWW%29%20%2A%20%5BDCN%20V2%5D%20DCN%20V2%20-%20Improved%20Deep%20%26%20Cross%20Network%20and%20Practical%20Lessons%20for%20Web-scale%20Learning%20to%20Rank%20Systems.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（Arxiv）** [ETA] 面向点击率预测的高效长序列用户数据建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2022%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BETA%5D%20Efficient%20Long%20Sequential%20User%20Data%20Modeling%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（WSDM）面向电商搜索点击率预测的用户情境化页面级反馈建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2022%20%28Alibaba%29%20%28WSDM%29%20Modeling%20Users%E2%80%99%20Contextualized%20Page-wise%20Feedback%20for%20Click-Through%20Rate%20Prediction%20in%20E-commerce%20Search.pdf) \u003Cbr \u002F>\n* [2022年（Meta）** （Arxiv）DHEN — 用于大规模点击率预测的深度分层集成网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2022%20%28Meta%29%20%2A%2A%20%28Arxiv%29%20DHEN%20-%20A%20Deep%20and%20Hierarchical%20Ensemble%20Network%20for%20Large-Scale%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（Arxiv）[ESLM] 全空间学习框架——在推荐系统全阶段实现无偏转化率预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Alibaba%29%20%28Arxiv%29%20%5BESLM%5D%20Entire%20Space%20Learning%20Framework%20-%20Unbias%20Conversion%20Rate%20Prediction%20in%20Full%20Stages%20of%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2023年（谷歌）（Arxiv）工厂车间——工业规模广告推荐模型的机器学习工程](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Google%29%20%28Arxiv%29%20On%20the%20Factory%20Floor%20-%20ML%20Engineering%20for%20Industrial-Scale%20Ads%20Recommendation%20Models.pdf) \u003Cbr \u002F>\n* [2023年（谷歌）** （Arxiv）[Hiformer] Hiformer — 基于Transformer的异构特征交互学习，用于推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Google%29%20%2A%2A%20%28Arxiv%29%20%5BHiformer%5D%20Hiformer%20-%20Heterogeneous%20Feature%20Interactions%20Learning%20with%20Transformers%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2023年（快手）（Arixiv）[TWIN] TWIN — 双阶段兴趣网络，用于快手点击率预测中的终身用户行为建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Kuaishou%29%20%28Arixiv%29%20%5BTWIN%5D%20TWIN%20-%20TWo-stage%20Interest%20Network%20for%20Lifelong%20User%20Behavior%20Modeling%20in%20CTR%20Prediction%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2023年（快手）（KDD）[PEPNet] PEPNet — 参数与嵌入个性化网络，用于注入个性化先验信息](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2023%20%28Kuaishou%29%20%28KDD%29%20%5BPEPNet%5D%20PEPNet%20-%20Parameter%20and%20Embedding%20Personalized%20Network%20for%20Infusing%20with%20Personalized%20Prior%20Information.pdf) \u003Cbr \u002F>\n* [2024年（快手）（CIKM）[TWINv2] TWIN V2 — 扩展超长用户行为序列建模，以提升快手点击率预测能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2024%20%28Kuaishou%29%20%28CIKM%29%20%5BTWINv2%5D%20TWIN%20V2%20-%20Scaling%20Ultra-Long%20User%20Behavior%20Sequence%20Modeling%20for%20Enhanced%20CTR%20Prediction%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2024年（Meta）（Arxiv）** [GR] 行动胜于言语——用于生成式推荐的万亿参数序列转换器](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2024%20%28Meta%29%20%28Arxiv%29%20%2A%2A%20%5BGR%5D%20Actions%20Speak%20Louder%20than%20Words%20-%20Trillion-Parameter%20Sequential%20Transducers%20for%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024年（Meta）** （PMLR）[Wukong] Wukong — 朝着大规模推荐的规模定律迈进](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2024%20%28Meta%29%20%2A%2A%20%28PMLR%29%20%5BWukong%5D%20Wukong%20-%20Towards%20a%20Scaling%20Law%20for%20Large-Scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）** （Arxiv）[LONGER] LONGER — 在工业级推荐系统中扩展长序列建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BLONGER%5D%20LONGER%20-%20Scaling%20Up%20Long%20Sequence%20Modeling%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）** （Arxiv）[RankMixer] RankMixer — 在工业级推荐系统中扩展排名模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）** （Arxiv）[STCA] 让它更长，保持快速——在抖音上以十亿规模实现端到端1万序列建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2025%20%EF%BC%88Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BSTCA%5D%20Make%20It%20Long%2C%20Keep%20It%20Fast%20-%20End-to-End%2010k-Sequence%20Modeling%20at%20Billion%20Scale%20on%20Douyin.pdf) \u003Cbr \u002F>\n* [2026年（Meta）（KDD）[Lattice] Meta Lattice — 面向成本效益的工业规模广告推荐的模型空间重新设计](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002F2026%20%28Meta%29%20%28KDD%29%20%5BLattice%5D%20Meta%20Lattice%20-%20Model%20Space%20Redesign%20for%20Cost-Effective%20Industry-Scale%20Ads%20Recommendations.pdf) \u003Cbr \u002F>\n\n#### 激活函数\n* [2020年（Arxiv） [GLU] GLU变体改进Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FActivation-Function\u002F2020%28Arxiv%29%20%20%5BGLU%5D%20GLU%20Variants%20Improve%20Transformer.pdf) \u003Cbr \u002F>\n\n#### 校准\n* [2014年（ADKDD）（Facebook）从Facebook广告点击预测中获得的实用经验](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FCalibration\u002F2014%20%28ADKDD%29%20%28Facebook%29%20Practical%20Lessons%20from%20Predicting%20Clicks%20on%20Ads%20at%20Facebook.pdf) \u003Cbr \u002F>\n* [2014年（TIST）展示广告的简单且可扩展的响应预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FCalibration\u002F2014%20%28TIST%29%20Simple%20and%20scalable%20response%20prediction%20for%20display%20advertising.pdf) \u003Cbr \u002F>\n* [2023年 ROC正则化等熵回归的分类器校准](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FCalibration\u002F2023%20Classifier%20Calibration%20with%20ROC-Regularized%20Isotonic%20Regression.pdf) \u003Cbr \u002F>\n\n#### 经典\n* [2016年（ICLR） [GRU4Rec] 基于会话的循环神经网络推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FClassic\u002F2016%20%28ICLR%29%20%5BGRU4Rec%5D%20Session-based%20Recommendations%20with%20Recurrent%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2017年（Amazon）（IEEE）亚马逊网站二十年的推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FClassic\u002F2017%20%28Amazon%29%20%28IEEE%29%20Two%20decades%20of%20recommender%20systems%20at%20Amazon.com.pdf) \u003Cbr \u002F>\n\n#### DNN\n* [2019年（KDD）（Airbnb）将深度学习应用于Airbnb搜索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDNN\u002F2019%20%28KDD%29%20%28Airbnb%29%20Applying%20Deep%20Learning%20To%20Airbnb%20Search.pdf) \u003Cbr \u002F>\n* [2020年（Airbnb）（KDD）改进Airbnb搜索的深度学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDNN\u002F2020%20%28Airbnb%29%20%28KDD%29%20Improving%20Deep%20Learning%20For%20Airbnb%20Search.pdf) \u003Cbr \u002F>\n\n#### 延迟反馈问题\n* [2008年（KDD）仅从正例和未标记数据中学习分类器](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2008%20%28KDD%29%20Learning%20Classifiers%20from%20Only%20Positive%20and%20Unlabeled%20Data.pdf) \u003Cbr \u002F>\n* [2014年（Criteo）（KDD）[DFM] 展示广告中的延迟反馈建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2014%20%28Criteo%29%20%28KDD%29%20%5BDFM%5D%20Modeling%20Delayed%20Feedback%20in%20Display%20Advertising.pdf) \u003Cbr \u002F>\n* [2018年（Arxiv）[NoDeF] 用于转化率预测的非参数化延迟反馈模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2018%20%28Arxiv%29%20%5BNoDeF%5D%20A%20Nonparametric%20Delayed%20Feedback%20Model%20for%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（Twitter）（RecSys）在CTR预测中使用神经网络进行持续训练时处理延迟反馈的问题](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2019%20%28Twitter%29%20%28RecSys%29%20Addressing%20Delayed%20Feedback%20for%20Continuous%20Training%20with%20Neural%20Networks%20in%20CTR%20prediction.pdf) \u003Cbr \u002F>\n* [2020年（AdKDD）基于负二项式回归的多转化延迟反馈模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2020%20%28AdKDD%29%20Delayed%20Feedback%20Model%20with%20Negative%20Binomial%20Regression%20for%20Multiple%20Conversions.pdf) \u003Cbr \u002F>\n* [2020年（京东）（IJCAI）[TS-DL] 基于注意力机制的延迟反馈转化率预测模型，通过点击后校准实现](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2020%20%28JD%29%20%28IJCAI%29%20%5BTS-DL%5D%20An%20Attention-based%20Model%20for%20Conversion%20Rate%20Prediction%20with%20Delayed%20Feedback%20via%20Post-click%20Calibration.pdf) \u003Cbr \u002F>\n* [2020年（SIGIR）[DLA-DF] 针对延迟转化的双学习算法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2020%20%28SIGIR%29%20%5BDLA-DF%5D%20Dual%20Learning%20Algorithm%20for%20Delayed%20Conversions.pdf) \u003Cbr \u002F>\n* [2020年（WWW）[FSIW] 在延迟反馈下预测转化率时的反馈偏移修正](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2020%20%28WWW%29%20%5BFSIW%5D%20A%20Feedback%20Shift%20Correction%20in%20Predicting%20Conversion%20Rates%20under%20Delayed%20Feedback.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（AAAI）[ES-DFM] 通过经过时间采样捕捉转化率预测中的延迟反馈](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Alibaba%29%20%28AAAI%29%20%5BES-DFM%5D%20Capturing%20Delayed%20Feedback%20in%20Conversion%20Rate%20Prediction%20via%20Elapsed-Time%20Sampling.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（AAAI）[ESDF] 针对全空间转化率预测的延迟反馈建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Alibaba%29%20%28AAAI%29%20%5BESDF%5D%20Delayed%20Feedback%20Modeling%20for%20the%20Entire%20Space%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（Arxiv）[Defer] 真实负样本很重要——使用真实负样本进行持续训练以建模延迟反馈](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Alibaba%29%20%28Arxiv%29%20%5BDefer%5D%20Real%20Negatives%20Matter%20-%20Continuous%20Training%20with%20Real%20Negatives%20for%20Delayed%20Feedback%20Modeling.pdf) \u003Cbr \u002F>\n* [2021年（谷歌）（Arxiv）在延迟反馈建模中处理每次点击的多次转化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Google%29%20%28Arxiv%29%20Handling%20many%20conversions%20per%20click%20in%20modeling%20delayed%20feedback.pdf) \u003Cbr \u002F>\n* [2021年（腾讯）（SIGIR）针对具有延迟反馈的流式推荐的反事实奖励修正](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2021%20%28Tencent%29%20%28SIGIR%29%20Counterfactual%20Reward%20Modification%20for%20Streaming%20Recommendation%20with%20Delayed%20Feedback.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（WWW）[DEFUSE] 通过标签修正实现延迟反馈建模的渐近无偏估计](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDelayed-Feedback-Problem\u002F2022%20%28Alibaba%29%20%28WWW%29%20%5BDEFUSE%5D%20Asymptotically%20Unbiased%20Estimation%20for%20Delayed%20Feedback%20Modeling%20via%20Label%20Correction.pdf) \u003Cbr \u002F>\n\n#### 蒸馏\n* [2020年（阿里巴巴）（KDD）*[特权特征蒸馏] 淘宝推荐中的特权特征蒸馏](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDistill\u002F2020%20%28Alibaba%29%20%28KDD%29%20%2A%5BPrivileged%20Features%20Distillation%5D%20Privileged%20Features%20Distillation%20at%20Taobao%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024年（谷歌）面向谷歌规模推荐系统的自辅助蒸馏，用于高效样本学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDistill\u002F2024%20%28Google%29%20Self-Auxiliary%20Distillation%20for%20Sample%20Efficient%20Learning%20in%20Google-Scale%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）（KDD）[HA-PFD] 基于潜在对齐的硬度感知特权特征蒸馏，用于CVR预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDistill\u002F2025%20%28Bytedance%29%20%28KDD%29%20%5BHA-PFD%5D%20Hardness-aware%20Privileged%20Features%20Distillation%20with%20Latent%20Alignment%20for%20CVR%20Prediction.pdf) \u003Cbr \u002F>\n* [2025年（快手）[MIKD] 面向短视频推荐的互信息感知知识蒸馏](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FDistill\u002F2025%20%28Kuaishou%29%20%5BMIKD%5D%20Mutual%20Information-aware%20Knowledge%20Distillation%20for%20Short%20Video%20Recommendation.pdf) \u003Cbr \u002F>\n\n#### 实验\n* [2010年（谷歌）重叠实验基础设施——更多、更好、更快的实验](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FExperiment\u002F2010%20%28Google%29%20Overlapping%20Experiment%20Infrastructure%20-%20More%2C%20Better%2C%20Faster%20Experimentation.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（KDD）淘宝展示广告中的优化每次点击成本](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FExperiment\u002F2019%20%28Alibaba%29%20%28KDD%29%20OptimizedCost%20perClickin%20TaobaoDisplayAdvertising.pdf) \u003Cbr \u002F>\n* [2022年（谷歌）（KDD）深度排序模型的规模校准](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FExperiment\u002F2022%20%28Google%29%20%28KDD%29%20Scale%20Calibration%20of%20Deep%20Ranking%20Models.pdf) \u003Cbr \u002F>\n\n#### 特征交叉\n* [2010年（ICDM）[FM] 因子分解机](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2010%20%28ICDM%29%20%5BFM%5D%20Factorization%20machines.pdf) \u003Cbr \u002F>\n* [2013年（谷歌）（KDD）[LR] 广告点击预测——来自一线的经验](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2013%20%28Google%29%20%28KDD%29%20%5BLR%5D%20Ad%20Click%20Prediction%20-%20a%20View%20from%20the%20Trenches.pdf) \u003Cbr \u002F>\n* [2016年（Arxiv）[PNN] 基于产品的神经网络用于用户响应预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28Arxiv%29%20%5BPNN%5D%20Product-based%20Neural%20Networks%20for%20User%20Response%20Prediction.pdf) \u003Cbr \u002F>\n* [2016年（Criteo）（Recsys）[FFM] 领域感知因子分解机用于CTR预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28Criteo%29%20%28Recsys%29%20%5BFFM%5D%20Field-aware%20Factorization%20Machines%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2016年（ECIR）[FNN] 多领域分类数据上的深度学习——以用户响应预测为例](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28ECIR%29%20%5BFNN%5D%20Deep%20Learning%20over%20Multi-field%20Categorical%20Data%20%E2%80%93%20A%20Case%20Study%20on%20User%20Response%20Prediction.pdf) \u003Cbr \u002F>\n* [2016年（KDD）[Deepintent] Deepintent——利用循环神经网络为在线广告学习注意力机制](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28KDD%29%20%5BDeepintent%5D%20Deepintent%20-%20Learning%20attentions%20for%20online%20advertising%20with%20recurrent%20neural%20networks.pdf) \u003Cbr \u002F>\n* [2016年（微软）（KDD）[Deep Crossing] Deep Crossing——无需人工设计组合特征的Web规模建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2016%20%28Microsoft%29%20%28KDD%29%20%5BDeep%20Crossing%5D%20Deep%20Crossing%20-%20Web-scale%20modeling%20without%20manually%20crafted%20combinatorial%20features.pdf) \u003Cbr \u002F>\n* [2017年（谷歌）（ADKDD）[DCN] 深度与交叉网络用于广告点击预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28Google%29%20%28ADKDD%29%20%5BDCN%5D%20Deep%20%26%20CrossNetwork%20for%20Ad%20Click%20Predictions.pdf) \u003Cbr \u002F>\n* [2017年（华为）（IJCAI）[DeepFM] DeepFM——基于因子分解机的神经网络用于CTR预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28Huawei%29%20%28IJCAI%29%20%5BDeepFM%5D%20DeepFM%20-%20A%20Factorization-Machine%20based%20Neural%20Network%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2017年（IJCAI）[AFM] 注意力因子分解机——通过注意力网络学习特征交互权重](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28IJCAI%29%20%5BAFM%5D%20Attentional%20Factorization%20Machines%20Learning%20the%20Weight%20of%20Feature%20Interactions%20via%20Attention%20Networks.pdf) \u003Cbr \u002F>\n* [2017年（SIGIR）[NFM] 用于稀疏预测分析的神经因子分解机](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28SIGIR%29%20%5BNFM%5D%20Neural%20Factorization%20Machines%20for%20Sparse%20Predictive%20Analytics.pdf) \u003Cbr \u002F>\n* [2017年（WWW）[NCF] 神经协同过滤](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2017%20%28WWW%29%20%5BNCF%5D%20Neural%20Collaborative%20Filtering.pdf) \u003Cbr \u002F>\n* [2018年（谷歌）（WSDM）[Latent Cross] Latent Cross——在循环推荐系统中利用上下文信息](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2018%20%28Google%29%20%28WSDM%29%20%5BLatent%20Cross%5D%20Latent%20Cross%20Making%20Use%20of%20Context%20in%20Recurrent%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2018年（KDD）[xDeepFM] xDeepFM——结合显式和隐式特征交互用于推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2018%20%28KDD%29%20%5BxDeepFM%5D%20xDeepFM%20-%20Combining%20Explicit%20and%20Implicit%20Feature%20Interactions%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2018年（TOIS）[PNN] 基于产品神经网络的多领域分类数据用户响应预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2018%20%28TOIS%29%20%5BPNN%5D%20Product-Based%20Neural%20Networks%20for%20User%20Response%20Prediction%20over%20Multi-Field%20Categorical%20Data.pdf) \u003Cbr \u002F>\n* [2019年（CIKM）** [AutoInt] AutoInt——通过自注意力神经网络自动学习特征交互](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2019%20%28CIKM%29%20%2A%2A%20%5BAutoInt%5D%20AutoInt%20-%20Automatic%20Feature%20Interaction%20Learning%20via%20Self-Attentive%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2019年（华为）（WWW）[FGCNN] 卷积神经网络生成特征用于点击率预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2019%20%28Huawei%29%20%28WWW%29%20%5BFGCNN%5D%20Feature%20Generation%20by%20Convolutional%20Neural%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（Sina）（Arxiv）[FAT-DeepFFM] FAT-DeepFFM——领域注意力深度领域感知因子分解机](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2019%20%28Sina%29%20%28Arxiv%29%20%5BFAT-DeepFFM%5D%20FAT-DeepFFM%20-%20Field%20Attentive%20Deep%20Field-aware%20Factorization%20Machine.pdf) \u003Cbr \u002F>\n* [2019年（腾讯）（AAAI）[IFM] 面向推荐系统的交互感知因子分解机](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2019%20%28Tencent%29%20%28AAAI%29%20%5BIFM%5D%20Interaction-aware%20Factorization%20Machines%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2020年（百度）（KDD）[CAN] 百度视频广告的组合注意力网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2020%20%28Baidu%29%20%28KDD%29%20%5BCAN%5D%20Combo-Attention%20Network%20for%20Baidu%20Video%20Advertising.pdf) \u003Cbr \u002F>\n* [2021年（谷歌）（NIPS）[MLP-Mixer] MLP-Mixer——一种全MLP架构用于视觉任务](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2021%20%28Google%29%20%28NIPS%29%20%5BMLP-Mixer%5D%20MLP-Mixer%20-%20An%20all-MLP%20Architecture%20for%20Vision.pdf) \u003Cbr \u002F>\n* [2021年（谷歌）（WWW）* [DCN V2] DCN V2——改进的深度与交叉网络及Web规模排序学习系统的实践经验](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2021%20%28Google%29%20%28WWW%29%20%2A%20%5BDCN%20V2%5D%20DCN%20V2%20-%20Improved%20Deep%20%26%20Cross%20Network%20and%20Practical%20Lessons%20for%20Web-scale%20Learning%20to%20Rank%20Systems.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（WSDM）* [CAN] CAN——用于点击率预测的特征协同作用网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2022%20%28Alibaba%29%20%28WSDM%29%20%2A%20%5BCAN%5D%20CAN%20-%20Feature%20Co-Action%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2022年（Meta）**（Arxiv）DHEN——用于大规模点击率预测的深度分层集成网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2022%20%28Meta%29%20%2A%2A%20%28Arxiv%29%20DHEN%20-%20A%20Deep%20and%20Hierarchical%20Ensemble%20Network%20for%20Large-Scale%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023年（CIKM）* [GDCN] 朝着更深层、更轻量且可解释的交叉网络发展，用于CTR预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2023%20%28CIKM%29%20%2A%20%5BGDCN%5D%20Towards%20Deeper%2C%20Lighter%20and%20Interpretable%20Cross%20Network%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2023年（谷歌）**（Arxiv）[Hiformer] Hiformer——利用Transformer学习异构特征交互，用于推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2023%20%28Google%29%20%2A%2A%20%28Arxiv%29%20%5BHiformer%5D%20Hiformer%20-%20Heterogeneous%20Feature%20Interactions%20Learning%20with%20Transformers%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2023年（Sina）（CIKM）[MemoNet] MemoNet——通过多哈希码本网络高效记忆所有交叉特征的表示，用于CTR预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2023%20%28Sina%29%20%28CIKM%29%20%5BMemoNet%5D%20MemoNet%20-%20Memorizing%20All%20Cross%20Features%E2%80%99%20Representations%20Efficiently%20via%20Multi-Hash%20Codebook%20Network%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2024年（Meta）**（PMLR）[Wukong] Wukong——迈向大规模推荐的规模定律](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2024%20%28Meta%29%20%2A%2A%20%28PMLR%29%20%5BWukong%5D%20Wukong%20-%20Towards%20a%20Scaling%20Law%20for%20Large-Scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024年（LinkedIn）（KDD）[RDCN] LiRank——LinkedIn的工业级大规模排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2024%20%EF%BC%88LinkedIn%29%20%28KDD%29%20%5BRDCN%5D%20LiRank%20-%20Industrial%20Large%20Scale%20Ranking%20Models%20at%20LinkedIn.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）[HHFT] HHFT——面向推荐系统的分层异构特征Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Alibaba%29%20%5BHHFT%5D%20HHFT%20-%20Hierarchical%20Heterogeneous%20Feature%20Transformer%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）**（Arxiv）[Pyramid Mixer] Pyramid Mixer——用于序列推荐的多维度多周期兴趣建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BPyramid%20Mixer%5D%20Pyramid%20Mixer%20-%20Multi-dimensional%20Multi-period%20Interest%20Modeling%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）**（Arxiv）[RankMixer] RankMixer——在工业推荐系统中扩展排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）**（CIKM）[RankMixer] RankMixer——在工业推荐系统中扩展排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Bytedance%29%20%2A%2A%20%28CIKM%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025年（Meta）（CIKM）[InterFormer] InterFormer——有效学习异构交互，用于点击率预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Meta%29%20%28CIKM%29%20%5BInterFormer%5D%20InterFormer%20-%20Effective%20Heterogeneous%20Interaction%20Learning%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2025年（腾讯）（Arxiv）[D-MoE] 通过去相关专家网络提升CTR预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Tencent%29%20%28Arxiv%29%20%5BD-MoE%5D%20Enhancing%20CTR%20Prediction%20with%20De-correlated%20Expert%20Networks.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）[FAT] 从规模化到结构化表达——重新思考用于CTR预测的Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%EF%BC%88Alibaba%29%20%5BFAT%5D%20From%20Scaling%20to%20Structured%20Expressivity%20-%20Rethinking%20Transformers%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）（Arxiv）[OneTrans] OneTrans——在工业推荐系统中用一个Transformer实现统一的特征交互与序列建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2025%20%28Bytedance%29%20%28Arxiv%29%20%5BOneTrans%5D%20OneTrans%20-%20Unified%20Feature%20Interaction%20and%20Sequence%20Modeling%20with%20One%20Transformer%20in%20Industrial%20Recommender.pdf) \u003Cbr \u002F>\n* [2026年（字节跳动）（Arxiv）[MixFormer] MixFormer——在工业推荐系统中同时扩展密集型和序列型模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BMixFormer%5D%20MixFormer%20-%20Co-Scaling%20Up%20Dense%20and%20Sequence%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026年（字节跳动）（Arxiv）[TokenMixer-Large] TokenMixer-Large——在工业推荐系统中扩展大型排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BTokenMixer-Large%5D%20TokenMixer-Large%20-%20Scaling%20Up%20Large%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026年（字节跳动）（Arxiv）[Zenith] Zenith——为十亿级直播推荐扩展排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BZenith%5D%20Zenith%20-%20Scaling%20up%20Ranking%20Models%20for%20Billion-scale%20Livestreaming%20Recommendation.pdf) \u003Cbr \u002F>\n* [2026年（快手）（Arxiv）[UniMixer] UniMixer——用于推荐系统规模定律的统一架构](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature-Crossing\u002F2026%20%28Kuaishou%29%20%28Arxiv%29%20%5BUniMixer%5D%20UniMixer%20-%20A%20Unified%20Architecture%20for%20Scaling%20Laws%20in%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n\n#### 特征重要性\n* [2022年（快手）（Arxiv）[LPFS] LPFS - 用于点击率预估的可学习极化特征选择](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature_Importance\u002F2022%20%28Kuaishou%29%20%28Arxiv%29%20%5BLPFS%5D%20LPFS%20-%20Learnable%20Polarizing%20Feature%20Selection%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2024年（华为）（KDD）ERASE - 面向深度推荐系统的特征选择方法基准测试](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FFeature_Importance\u002F2024%20%28Huawei%29%28KDD%29%20ERASE%20-%20Benchmarking%20Feature%20Selection%20Methods%20for%20Deep%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n\n#### 门控机制\n* [2014年（TASLP）* [LHUC] 用于无监督声学模型自适应的隐藏单元贡献学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2014%20%28TASLP%29%20%2A%20%5BLHUC%5D%20Learning%20Hidden%20Unit%20Contributions%20for%20Unsupervised%20Acoustic%20Model%20Adaptation.pdf) \u003Cbr \u002F>\n* [2018年（CVPR）* [SENet] 激励挤压网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2018%20%28CVPR%29%20%2A%20%5BSENet%5D%20Squeeze-and-Excitation%20Networks.pdf) \u003Cbr \u002F>\n* [2019年（新浪）（Recsys）[FiBiNET] FiBiNET - 结合特征重要性和双线性特征交互的点击率预估](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2019%20%28Sina%29%20%28Recsys%29%20%5BFiBiNET%5D%20FiBiNET%20-%20combining%20feature%20importance%20and%20bilinear%20feature%20interaction%20for%20click-through%20rate%20prediction.pdf) \u003Cbr \u002F>\n* [2020年（新浪）（Arxiv）[GateNet] GateNet - 用于点击率预估的门控增强深度网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2020%20%28Sina%29%20%28Arxiv%29%20%5BGateNet%5D%20GateNet%20-%20Gating-Enhanced%20Deep%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2021年（新浪）（Arxiv）[ContextNet] ContextNet - 利用上下文信息精炼特征嵌入的点击率预估框架](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2021%20%28Sina%29%20%28Arxiv%29%20%5BContextNet%5D%20ContextNet%20-%20A%20Click-Through%20Rate%20Prediction%20Framework%20Using%20Contextual%20information%20to%20Refine%20Feature%20Embedding.pdf) \u003Cbr \u002F>\n* [2021年（新浪）（DLP-KDD）[MaskNet] MaskNet - 通过实例引导的掩码将逐特征乘法引入CTR排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2021%20%28Sina%29%20%EF%BC%88DLP-KDD%29%20%5BMaskNet%5D%20MaskNet%20-%20Introducing%20Feature-Wise%20Multiplication%20to%20CTR%20Ranking%20Models%20by%20Instance-Guided%20Mask.pdf) \u003Cbr \u002F>\n* [2023年（快手）（KDD）[PEPNet] PEPNet - 用于注入个性化先验信息的参数与嵌入个性化网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2023%20%28Kuaishou%29%20%28KDD%29%20%5BPEPNet%5D%20PEPNet%20-%20Parameter%20and%20Embedding%20Personalized%20Network%20for%20Infusing%20with%20Personalized%20Prior%20Information.pdf) \u003Cbr \u002F>\n* [2023年（新浪）（CIKM）[FiBiNet++] FiBiNet++ - 通过低秩特征交互层降低CTR预测模型规模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2023%20%28Sina%29%20%28CIKM%29%20%5BFiBiNet%2B%2B%5D%20FiBiNet%2B%2B%20-%20Reducing%20Model%20Size%20by%20Low%20Rank%20Feature%20Interaction%20Layer%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）[ADS] 推荐系统中面向个性化序列建模的自适应域缩放](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FGating\u002F2025%20%28Bytedance%29%20%5BADS%5D%20Adaptive%20Domain%20Scaling%20for%20Personalized%20Sequential%20Modeling%20in%20Recommenders.pdf) \u003Cbr \u002F>\n\n#### LLM_Ranking\n* [2019 (CIKM)  [AutoInt] AutoInt -Automatic Feature Interaction Learning via Self-Attentive Neural Networks](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2019%20%28CIKM%29%20%20%5BAutoInt%5D%20AutoInt%20-Automatic%20Feature%20Interaction%20Learning%20via%20Self-Attentive%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2020 (Arxiv) Scaling Laws for Neural Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2020%20%28Arxiv%29%20Scaling%20Laws%20for%20Neural%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2021 (Baidu) (KDD) Pre-trained Language Model based Ranking in Baidu Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2021%20%28Baidu%29%20%28KDD%29%20Pre-trained%20Language%20Model%20based%20Ranking%20in%20Baidu%20Search.pdf) \u003Cbr \u002F>\n* [2021 (Google) (Arxiv) [MLP-Mixer] MLP-Mixer - An all-MLP Architecture for Vision](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2021%20%28Google%29%20%28Arxiv%29%20%5BMLP-Mixer%5D%20MLP-Mixer%20-%20An%20all-MLP%20Architecture%20for%20Vision.pdf) \u003Cbr \u002F>\n* [2022 (Meta) ** (Arxiv) DHEN - A Deep and Hierarchical Ensemble Network for Large-Scale Click-Through Rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2022%20%28Meta%29%20%2A%2A%20%28Arxiv%29%20DHEN%20-%20A%20Deep%20and%20Hierarchical%20Ensemble%20Network%20for%20Large-Scale%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023 (Arxiv) [E4SRec] E4SRec - An Elegant Effective Efficient Extensible Solution of Large Language Models for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2023%20%28Arxiv%29%20%5BE4SRec%5D%20E4SRec%20-%20An%20Elegant%20Effective%20Efficient%20Extensible%20Solution%20of%20Large%20Language%20Models%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023 (Google) ** (Arxiv) [Hiformer] Hiformer - Heterogeneous Feature Interactions Learning with Transformers for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2023%20%28Google%29%20%2A%2A%20%28Arxiv%29%20%5BHiformer%5D%20Hiformer%20-%20Heterogeneous%20Feature%20Interactions%20Learning%20with%20Transformers%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2024 (Alibaba) (Arxiv) [BAHE] Breaking the Length Barrier - LLM-Enhanced CTR Prediction in Long Textual User Behaviors](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Alibaba%29%20%28Arxiv%29%20%5BBAHE%5D%20Breaking%20the%20Length%20Barrier%20-%20LLM-Enhanced%20CTR%20Prediction%20in%20Long%20Textual%20User%20Behaviors.pdf) \u003Cbr \u002F>\n* [2024 (Bytedance) (Arxiv) [HLLM] HLLM - Enhancing Sequential Recommendations via Hierarchical Large Language Models for Item and User Modeling](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Bytedance%29%20%28Arxiv%29%20%5BHLLM%5D%20HLLM%20-%20Enhancing%20Sequential%20Recommendations%20via%20Hierarchical%20Large%20Language%20Models%20for%20Item%20and%20User%20Modeling.pdf) \u003Cbr \u002F>\n* [2024 (Google) (Arxiv) LLMs for User Interest Exploration in Large-scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Google%29%20%28Arxiv%29%20LLMs%20for%20User%20Interest%20Exploration%20in%20Large-scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2024 (Google) (Arxiv) [CALRec] CALRec - Contrastive Alignment of Generative LLMs for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Google%29%20%28Arxiv%29%20%5BCALRec%5D%20CALRec%20-%20Contrastive%20Alignment%20of%20Generative%20LLMs%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 (Google) (ICLR) From Sparse to Soft Mixtures of Experts](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Google%29%20%28ICLR%29%20From%20Sparse%20to%20Soft%20Mixtures%20of%20Experts.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (Arxiv) [LEARN] LEARN - Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Application](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Kuaishou%29%20%28Arxiv%29%20%5BLEARN%5D%20LEARN%20-%20Knowledge%20Adaptation%20from%20Large%20Language%20Model%20to%20Recommendation%20for%20Practical%20Industrial%20Application.pdf) \u003Cbr \u002F>\n* [2024 (Kuaishou) (KDD) [NAR4Rec] Non-autoregressive Generative Models for Reranking Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Kuaishou%29%20%28KDD%29%20%5BNAR4Rec%5D%20Non-autoregressive%20Generative%20Models%20for%20Reranking%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 (Meituan) (Arxiv) [SRP4CTR] Enhancing CTR Prediction through Sequential Recommendation Pre-training - Introducing the SRP4CTR Framework](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meituan%29%20%28Arxiv%29%20%5BSRP4CTR%5D%20Enhancing%20CTR%20Prediction%20through%20Sequential%20Recommendation%20Pre-training%20-%20Introducing%20the%20SRP4CTR%20Framework.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) ** [GR] Actions Speak Louder than Words - Trillion-Parameter Sequential Transducers for Generative Recommendations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meta%29%20%28Arxiv%29%20%2A%2A%20%5BGR%5D%20Actions%20Speak%20Louder%20than%20Words%20-%20Trillion-Parameter%20Sequential%20Transducers%20for%20Generative%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) Unifying Generative and Dense Retrieval for Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meta%29%20%28Arxiv%29%20Unifying%20Generative%20and%20Dense%20Retrieval%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024 (Meta) (Arxiv) [SUM] Scaling User Modeling - Large-scale Online User Representations for Ads Personalization in Meta](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meta%29%20%28Arxiv%29%20%5BSUM%5D%20Scaling%20User%20Modeling%20-%20Large-scale%20Online%20User%20Representations%20for%20Ads%20Personalization%20in%20Meta.pdf) \u003Cbr \u002F>\n* [2024 (Meta) ** (PMLR) [Wukong] Wukong - Towards a Scaling Law for Large-Scale Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2024%20%28Meta%29%20%2A%2A%20%28PMLR%29%20%5BWukong%5D%20Wukong%20-%20Towards%20a%20Scaling%20Law%20for%20Large-Scale%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025  (Kuaishou) (Arxiv)[OneRec] OneRec - Unifying Retrieve and Rank with Generative Recommender and Preference Alignment](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%20%28Kuaishou%29%20%28Arxiv%29%5BOneRec%5D%20OneRec%20-%20Unifying%20Retrieve%20and%20Rank%20with%20Generative%20Recommender%20and%20Preference%20Alignment.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) Unlocking Scaling Law in Industrial Recommendation Systems with a Three-step Paradigm based Large User Model](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20Unlocking%20Scaling%20Law%20in%20Industrial%20Recommendation%20Systems%20with%20a%20Three-step%20Paradigm%20based%20Large%20User%20Model.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [HeterRec] Hierarchical Causal Transformer with Heterogeneous Information for Expandable Sequential Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BHeterRec%5D%20Hierarchical%20Causal%20Transformer%20with%20Heterogeneous%20Information%20for%20Expandable%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [LREA] Efficient Long Sequential Low-rank Adaptive Attention for Click-through rate Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BLREA%5D%20Efficient%20Long%20Sequential%20Low-rank%20Adaptive%20Attention%20for%20Click-through%20rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (Arxiv) [URM] Large Language Models Are Universal Recommendation Learners](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BURM%5D%20Large%20Language%20Models%20Are%20Universal%20Recommendation%20Learners.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (KDD) [GPSD] Scaling Transformers for Discriminative Recommendation via Generative Pretraining](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28KDD%29%20%5BGPSD%5D%20Scaling%20Transformers%20for%20Discriminative%20Recommendation%20via%20Generative%20Pretraining.pdf) \u003Cbr \u002F>\n* [2025 (Alibaba) (WWW) Explainable LLM-driven Multi-dimensional Distillation for E-Commerce Relevance Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Alibaba%29%20%28WWW%29%20Explainable%20LLM-driven%20Multi-dimensional%20Distillation%20for%20E-Commerce%20Relevance%20Learning.pdf) \u003Cbr \u002F>\n* [2025 (Amazon) (Arxiv) SynerGen - Contextualized Generative Recommender for Unified Search and Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Amazon%29%20%28Arxiv%29%20SynerGen%20-%20Contextualized%20Generative%20Recommender%20for%20Unified%20Search%20and%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Arxiv) (Pinterest) [PinRec] PinRec - Outcome-Conditioned, Multi-Token Generative Retrieval for Industry-Scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Arxiv%29%20%28Pinterest%29%20%5BPinRec%5D%20PinRec%20-%20Outcome-Conditioned%2C%20Multi-Token%20Generative%20Retrieval%20for%20Industry-Scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Arxiv) (Xiaohongshu) [GenRank] Towards Large-scale Generative Ranking](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Arxiv%29%20%28Xiaohongshu%29%20%5BGenRank%5D%20Towards%20Large-scale%20Generative%20Ranking.pdf) \u003Cbr \u002F>\n* [2025 (Baidu) (Arxiv) [COBRA] Sparse Meets Dense -Unified Generative Recommendations with Cascaded Sparse-Dense Representations](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Baidu%29%20%28Arxiv%29%20%5BCOBRA%5D%20Sparse%20Meets%20Dense%20-Unified%20Generative%20Recommendations%20with%20Cascaded%20Sparse-Dense%20Representations.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) (Arxiv) [RankMixer] RankMixer - Scaling Up Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Bytedance%29%20%28Arxiv%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (Arxiv) [LONGER] LONGER - Scaling Up Long Sequence Modeling in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BLONGER%5D%20LONGER%20-%20Scaling%20Up%20Long%20Sequence%20Modeling%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Bytedance) ** (CIKM) [RankMixer] RankMixer - Scaling Up Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Bytedance%29%20%2A%2A%20%28CIKM%29%20%5BRankMixer%5D%20RankMixer%20-%20Scaling%20Up%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025 (Google) (Arxiv) User Feedback Alignment for LLM-powered Exploration in Large-scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Google%29%20%28Arxiv%29%20User%20Feedback%20Alignment%20for%20LLM-powered%20Exploration%20in%20Large-scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Google) (Arxiv) [STAR] STAR - A Simple Training-free Approach for Recommendations using Large Language Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Google%29%20%28Arxiv%29%20%5BSTAR%5D%20STAR%20-%20A%20Simple%20Training-free%20Approach%20for%20Recommendations%20using%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2025 (Google) ** (Arxiv) [Hiformer] Hiformer - Heterogeneous Feature Interactions Learning with Transformers for Recommender Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Google%29%20%2A%2A%20%28Arxiv%29%20%5BHiformer%5D%20Hiformer%20-%20Heterogeneous%20Feature%20Interactions%20Learning%20with%20Transformers%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [GenSAR] Unified Generative Search and Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BGenSAR%5D%20Unified%20Generative%20Search%20and%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [LARM] LLM-Alignment Live-Streaming Recommendationpdf](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BLARM%5D%20LLM-Alignment%20Live-Streaming%20Recommendationpdf.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [LEARN] LEARN - Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Application](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BLEARN%5D%20LEARN%20-%20Knowledge%20Adaptation%20from%20Large%20Language%20Model%20to%20Recommendation%20for%20Practical%20Industrial%20Application.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneLoc] OneLoc - Geo-Aware Generative Recommender Systems for Local Life Service](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneLoc%5D%20OneLoc%20-%20Geo-Aware%20Generative%20Recommender%20Systems%20for%20Local%20Life%20Service.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneRec-V2] OneRec Technical Report v2](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneRec-V2%5D%20OneRec%20Technical%20Report%20v2.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneRec] OneRec - Unifying Retrieve and Rank with Generative Recommender and Preference Alignment](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneRec%5D%20OneRec%20-%20Unifying%20Retrieve%20and%20Rank%20with%20Generative%20Recommender%20and%20Preference%20Alignment.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneRec] OneRec Technical Report](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneRec%5D%20OneRec%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneSearch] OneSearch - A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneSearch%5D%20OneSearch%20-%20A%20Preliminary%20Exploration%20of%20the%20Unified%20End-to-End%20Generative%20Framework%20for%20E-commerce%20Search.pdf) \u003Cbr \u002F>\n* [2025 (Kuaishou) (Arxiv) [OneSug] OneSug - The Unified End-to-End Generative Framework for E-commerce Query Suggestion](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BOneSug%5D%20OneSug%20-%20The%20Unified%20End-to-End%20Generative%20Framework%20for%20E-commerce%20Query%20Suggestion.pdf) \u003Cbr \u002F>\n* [2025 (Meituan) (Arxiv) [DFGR] Action is All You Need - Dual-Flow Generative Ranking Network for Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Meituan%29%20%28Arxiv%29%20%5BDFGR%5D%20Action%20is%20All%20You%20Need%20-%20Dual-Flow%20Generative%20Ranking%20Network%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (Meituan) (Arxiv) [MTGR] MTGR - Industrial-Scale Generative Recommendation Framework in Meituan](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Meituan%29%20%28Arxiv%29%20%5BMTGR%5D%20MTGR%20-%20Industrial-Scale%20Generative%20Recommendation%20Framework%20in%20Meituan.pdf) \u003Cbr \u002F>\n* [2025 (Meituan) (Arxiv) [UniROM] One Model to Rank Them All - Unifying Online Advertising with End-to-End Learning](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Meituan%29%20%28Arxiv%29%20%5BUniROM%5D%20One%20Model%20to%20Rank%20Them%20All%20-%20Unifying%20Online%20Advertising%20with%20End-to-End%20Learning.pdf) \u003Cbr \u002F>\n* [2025 (Meta) (Arxiv) [HyperCast] Realizing Scaling Laws in Recommender Systems - A Foundation–Expert Paradigm for Hyperscale Model Deployment](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Meta%29%20%28Arxiv%29%20%5BHyperCast%5D%20Realizing%20Scaling%20Laws%20in%20Recommender%20Systems%20-%20A%20Foundation%E2%80%93Expert%20Paradigm%20for%20Hyperscale%20Model%20Deployment.pdf) \u003Cbr \u002F>\n* [2025 (Microsoft) (KDD)Towards Web-scale Recommendations with LLMs - From Quality-aware Ranking to Candidate Generation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Microsoft%29%20%28KDD%29Towards%20Web-scale%20Recommendations%20with%20LLMs%20-%20From%20Quality-aware%20Ranking%20to%20Candidate%20Generation.pdf) \u003Cbr \u002F>\n* [2025 (Pinterest) (Arxiv) [PinFM] PinFM - Foundation Model for User Activity Sequences at a Billion-scale Visual Discovery Platform](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Pinterest%29%20%28Arxiv%29%20%5BPinFM%5D%20PinFM%20-%20Foundation%20Model%20for%20User%20Activity%20Sequences%20at%20a%20Billion-scale%20Visual%20Discovery%20Platform.pdf) \u003Cbr \u002F>\n* [2025 (Shopee) （Arxiv) OnePiece - Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Shopee%29%20%EF%BC%88Arxiv%29%20OnePiece%20-%20Bringing%20Context%20Engineering%20and%20Reasoning%20to%20Industrial%20Cascade%20Ranking%20System.pdf) \u003Cbr \u002F>\n* [2025 (Tencent) (Arxiv) [GPR] GPR - Towards a Generative Pre-trained One-Model Paradigm for Large-Scale Advertising Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28Tencent%29%20%28Arxiv%29%20%5BGPR%5D%20GPR%20-%20Towards%20a%20Generative%20Pre-trained%20One-Model%20Paradigm%20for%20Large-Scale%20Advertising%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025 (eBay)  (Arxiv) LLMDistill4Ads - Using Cross-Encoders to Distill from LLM Signals for Advertiser Keyphrase Recommendations at eBay](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%28eBay%29%20%20%28Arxiv%29%20LLMDistill4Ads%20-%20Using%20Cross-Encoders%20to%20Distill%20from%20LLM%20Signals%20for%20Advertiser%20Keyphrase%20Recommendations%20at%20eBay.pdf) \u003Cbr \u002F>\n* [2025 （Bytedance) (Arxiv) [OneTrans] OneTrans - Unified Feature Interaction and Sequence Modeling with One Transformer in Industrial Recommender](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%EF%BC%88Bytedance%29%20%28Arxiv%29%20%5BOneTrans%5D%20OneTrans%20-%20Unified%20Feature%20Interaction%20and%20Sequence%20Modeling%20with%20One%20Transformer%20in%20Industrial%20Recommender.pdf) \u003Cbr \u002F>\n* [2025 （Bytedance) ** (Arxiv) [STCA] Make It Long, Keep It Fast - End-to-End 10k-Sequence Modeling at Billion Scale on Douyin](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2025%20%EF%BC%88Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BSTCA%5D%20Make%20It%20Long%2C%20Keep%20It%20Fast%20-%20End-to-End%2010k-Sequence%20Modeling%20at%20Billion%20Scale%20on%20Douyin.pdf) \u003Cbr \u002F>\n* [2026 (Alibaba) (Arxiv) [EST] EST - Towards Efficient Scaling Laws in Click-Through Rate Prediction via Unified Modeling](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Alibaba%29%20%28Arxiv%29%20%5BEST%5D%20EST%20-%20Towards%20Efficient%20Scaling%20Laws%20in%20Click-Through%20Rate%20Prediction%20via%20Unified%20Modeling.pdf) \u003Cbr \u002F>\n* [2026 (Alibaba) (Arxiv) [SORT] SORT - A Systematically Optimized Ranking Transformer for Industrial-scale Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Alibaba%29%20%28Arxiv%29%20%5BSORT%5D%20SORT%20-%20A%20Systematically%20Optimized%20Ranking%20Transformer%20for%20Industrial-scale%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [HyFormer] HyFormer - Revisiting the Roles of Sequence Modeling and Feature Interaction in CTR Prediction](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BHyFormer%5D%20HyFormer%20-%20Revisiting%20the%20Roles%20of%20Sequence%20Modeling%20and%20Feature%20Interaction%20in%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [MixFormer] MixFormer - Co-Scaling Up Dense and Sequence in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BMixFormer%5D%20MixFormer%20-%20Co-Scaling%20Up%20Dense%20and%20Sequence%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [TokenMixer-Large] TokenMixer-Large - Scaling Up Large Ranking Models in Industrial Recommenders](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BTokenMixer-Large%5D%20TokenMixer-Large%20-%20Scaling%20Up%20Large%20Ranking%20Models%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [UG-Sep] Compute Only Once - UG-Separation for Efficient Large Recommendation Models](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BUG-Sep%5D%20Compute%20Only%20Once%20-%20UG-Separation%20for%20Efficient%20Large%20Recommendation%20Models.pdf) \u003Cbr \u002F>\n* [2026 (Bytedance) (Arxiv) [Zenith] Zenith - Scaling up Ranking Models for Billion-scale Livestreaming Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Bytedance%29%20%28Arxiv%29%20%5BZenith%5D%20Zenith%20-%20Scaling%20up%20Ranking%20Models%20for%20Billion-scale%20Livestreaming%20Recommendation.pdf) \u003Cbr \u002F>\n* [2026 (Kuaishou) (Arxiv) [GR4AD] Generative Recommendation for Large-Scale Advertising](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Kuaishou%29%20%28Arxiv%29%20%5BGR4AD%5D%20Generative%20Recommendation%20for%20Large-Scale%20Advertising.pdf) \u003Cbr \u002F>\n* [2026 (Meta) (Arxiv) [Kunlun] Kunlun - Establishing Scaling Laws for Massive-Scale Recommendation Systems through Unified Architecture Design](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Meta%29%20%28Arxiv%29%20%5BKunlun%5D%20Kunlun%20-%20Establishing%20Scaling%20Laws%20for%20Massive-Scale%20Recommendation%20Systems%20through%20Unified%20Architecture%20Design.pdf) \u003Cbr \u002F>\n* [2026 (Meta) (Arxiv) [LLaTTE] LLaTTE - Scaling Laws for Multi-Stage Sequence Modeling in Large-Scale Ads Recommendation](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Meta%29%20%28Arxiv%29%20%5BLLaTTE%5D%20LLaTTE%20-%20Scaling%20Laws%20for%20Multi-Stage%20Sequence%20Modeling%20in%20Large-Scale%20Ads%20Recommendation.pdf) \u003Cbr \u002F>\n* [2026 (Meta) (Arxiv) [ULTRA-HSTU] Bending the Scaling Law Curve in Large-Scale Recommendation Systems](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%28Meta%29%20%28Arxiv%29%20%5BULTRA-HSTU%5D%20Bending%20the%20Scaling%20Law%20Curve%20in%20Large-Scale%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2026 （Bytedance) (Arxiv) [MDL] MDL - A Unified Multi-Distribution Learner in Large-scale Industrial Recommendation through Tokenization](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002F2026%20%EF%BC%88Bytedance%29%20%28Arxiv%29%20%5BMDL%5D%20MDL%20-%20A%20Unified%20Multi-Distribution%20Learner%20in%20Large-scale%20Industrial%20Recommendation%20through%20Tokenization.pdf) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLLM_Ranking\u002FLLM_MultiModal_Ranking) \u003Cbr \u002F>\n\n#### 损失函数\n* [2015年（Twitter）（KDD）Twitter时间线广告点击率预估](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2015%20%28Twitter%29%20%28KDD%29%20Click-through%20Prediction%20for%20Advertising%20in%20Twitter%20Timeline.pdf) \u003Cbr \u002F>\n* [2022年（Google）（KDD）深度排序模型的规模校准](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2022%20%28Google%29%20%28KDD%29%20Scale%20Calibration%20of%20Deep%20Ranking%20Models.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（KDD）基于上下文混合模型的排序与校准联合优化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2023%20%28Alibaba%29%20%28KDD%29%20Joint%20Optimization%20of%20Ranking%20and%20Calibration%20with%20Contextualized%20Hybrid%20Model.pdf) \u003Cbr \u002F>\n* [2023年（Google）（CIKM）面向二值相关性的校准排序的回归兼容列表级目标函数](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2023%20%28Google%29%20%28CIKM%29%20Regression%20Compatible%20Listwise%20Objectives%20for%20Calibrated%20Ranking%20with%20Binary%20Relevance.pdf) \u003Cbr \u002F>\n* [2024年（腾讯）（KDD）理解稀疏用户反馈下的推荐排序损失](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FLoss\u002F2024%20%28Tencent%29%20%28KDD%29%20Understanding%20the%20Ranking%20Loss%20for%20Recommendation%20with%20Sparse%20User%20Feedback.pdf) \u003Cbr \u002F>\n\n#### 多模态\n* [2018年（阿里巴巴）（CIKM）[图片CTR] 图片很重要——利用先进模型服务器对用户行为进行视觉建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-Modal\u002F2018%20%28Alibaba%29%20%28CIKM%29%20%5BImage%20CTR%5D%20Image%20Matters%20-%20Visually%20Modeling%20User%20Behaviors%20Using%20Advanced%20Model%20Server.pdf) \u003Cbr \u002F>\n* [2024年（阿里巴巴）（CIKM）利用多模态表示增强淘宝展示广告——挑战、方法与洞察](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-Modal\u002F2024%20%28Alibaba%29%20%28CIKM%29%20Enhancing%20Taobao%20Display%20Advertising%20with%20Multimodal%20Representations%20-%20Challenges%2C%20Approaches%20and%20Insights.pdf) \u003Cbr \u002F>\n* [2026年（阿里巴巴）（WSDM）[MOON] MOON——基于生成式多模态大语言模型的多模态表征学习，用于电商商品理解](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-Modal\u002F2026%20%28Alibaba%29%20%28WSDM%29%20%5BMOON%5D%20MOON%20-%20Generative%20MLLM-based%20Multimodal%20Representation%20Learning%20for%20E-commerce%20Product%20Understanding.pdf) \u003Cbr \u002F>\n\n#### 多领域-多场景\n* [2014年（TASLP） * [LHUC] 用于无监督声学模型自适应的隐藏单元贡献学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2014%20%28TASLP%29%20%2A%20%5BLHUC%5D%20Learning%20Hidden%20Unit%20Contributions%20for%20Unsupervised%20Acoustic%20Model%20Adaptation.pdf) \u003Cbr \u002F>\n* [2015年（微软）（WWW）推荐系统中跨域用户建模的多视图深度学习方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2015%20%28Microsoft%29%20%28WWW%29%20A%20Multi-View%20Deep%20Learning%20Approach%20for%20Cross%20Domain%20User%20Modeling%20in%20Recommendation%20Systems.pdf) \u003Cbr \u002F>\n* [2018年（谷歌）（KDD） ** [MMoE] 基于多门控专家混合的多任务学习中任务关系建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2018%20%28Google%29%20%28KDD%29%20%2A%2A%20%5BMMoE%5D%20Modeling%20task%20relationships%20in%20multi-task%20learning%20with%20multi-gate%20mixture-of-experts.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（CIKM） [WE-CAN] 带有Wasserstein正则项的跨域注意力网络用于电商搜索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2019%20%28Alibaba%29%20%28CIKM%29%20%5BWE-CAN%5D%20Cross-domain%20Attention%20Network%20with%20Wasserstein%20Regularizers%20for%20E-commerce%20Search.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（Arxiv） [SAML] 面向电商多场景推荐的情境感知与互惠驱动方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2020%20%28Alibaba%29%20%28Arxiv%29%20%5BSAML%5D%20Scenario-aware%20and%20Mutual-based%20approach%20for%20Multi-scenario%20Recommendation%20in%20E-Commerce.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（CIKM） [HMoE] 通过利用标签空间中的任务关系提升电商多场景排序学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2020%20%28Alibaba%29%20%28CIKM%29%20%5BHMoE%5D%20Improving%20Multi-Scenario%20Learning%20to%20Rank%20in%20E-commerce%20by%20Exploiting%20Task%20Relationships%20in%20the%20Label%20Space.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（CIKM） [MiNet] MiNet - 用于跨域点击率预估的混合兴趣网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2020%20%28Alibaba%29%28CIKM%29%20%5BMiNet%5D%20MiNet%20-%20Mixed%20Interest%20Network%20for%20Cross-Domain%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020年（腾讯）（Recsys） **  [PLE] 渐进式分层提取（PLE）——一种用于个性化推荐的新型多任务学习（MTL）模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2020%20%28Tencent%29%20%28Recsys%29%20%2A%2A%20%5BPLE%5D%20Progressive%20Layered%20Extraction%20%28PLE%29%20-%20A%20Novel%20Multi-Task%20Learning%20%28MTL%29%20Model%20for%20Personalized%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（CIKM） * [ZEUS] 针对电商多场景排序的用户自发行为自监督学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%2A%20%5BZEUS%5D%20Self-Supervised%20Learning%20on%20Users%E2%80%99%20Spontaneous%20Behaviors%20for%20Multi-Scenario%20Ranking%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（CIKM） ** [STAR] 一模型通天下——面向多域点击率预估的星型拓扑自适应推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%2A%2A%20%5BSTAR%5D%20One%20Model%20to%20Serve%20All%20-%20Star%20Topology%20Adaptive%20Recommender%20for%20Multi-Domain%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2021年（谷歌）（ICLR） 超网格Transformer——迈向多任务单一模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2021%20%28Google%29%20%28ICLR%29%20HyperGrid%20Transformers%20-%20Towards%20A%20Single%20Model%20for%20Multiple%20Tasks.pdf) \u003Cbr \u002F>\n* [2021年（快手）（Arxiv） [POSO] POSO——大规模推荐系统中的个性化冷启动模块](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2021%20%28Kwai%29%20%28Arxiv%29%20%5BPOSO%5D%20POSO%20-%20Personalized%20Cold%20Start%20Modules%20for%20Large-scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（CIKM） AdaSparse——面向多域点击率预估的自适应稀疏结构学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2022%20%28Alibaba%29%20%28CIKM%29%20AdaSparse%20-%20Learning%20Adaptively%20Sparse%20Structures%20for%20Multi-Domain%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（NIPS） ** [APG] APG——用于点击率预估的自适应参数生成网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2022%20%28Alibaba%29%20%28NIPS%29%20%2A%2A%20%5BAPG%5D%20APG%20-%20Adaptive%20Parameter%20Generation%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（CIKM） [HC2] 用于多场景广告排序的混合对比约束](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BHC2%5D%20Hybrid%20Contrastive%20Constraints%20for%20Multi-Scenario%20Ad%20Ranking.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（CIKM） [MMN] 掩码多域网络——单模型实现多类型、多场景转化率预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BMMN%5D%20Masked%20Multi-Domain%20Network%20-%20Multi-Type%20and%20Multi-Scenario%20Conversion%20Rate%20Prediction%20with%20a%20Single%20Model.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（CIKM） [Rec4Ad] Rec4Ad——淘宝广告点击率预估中缓解样本选择偏倚的免费午餐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BRec4Ad%5D%20Rec4Ad%20-%20A%20Free%20Lunch%20to%20Mitigate%20Sample%20Selection%20Bias%20for%20Ads%20CTR%20Prediction%20in%20Taobao.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（SIGIR） [MARIA] 基于自适应特征学习的多场景排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Alibaba%29%20%28SIGIR%29%20%5BMARIA%5D%20Multi-Scenario%20Ranking%20with%20Adaptive%20Feature%20Learning.pdf) \u003Cbr \u002F>\n* [2023年（CIKM） [HAMUR] HAMUR——多域推荐的超适配器](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28CIKM%29%20%5BHAMUR%5D%20HAMUR%20-%20Hyper%20Adapter%20for%20Multi-Domain%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023年（华为）（CIKM） [DFFM] DFFM——面向点击率预估的领域促进特征建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Huawei%29%20%28CIKM%29%20%5BDFFM%5D%20DFFM%20-%20Domain%20Facilitated%20Feature%20Modeling%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2023年（快手）（KDD） *  [PEPNet] PEPNet——用于注入个性化先验信息的参数与嵌入个性化网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Kuaishou%29%20%28KDD%29%20%2A%20%5BPEPNet%5D%20PEPNet%20-%20Parameter%20and%20Embedding%20Personalized%20Network%20for%20Infusing%20with%20Personalized%20Prior%20Information.pdf) \u003Cbr \u002F>\n* [2023年（腾讯）（KDD） 场景适应性特征交互用于点击率预估](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2023%20%28Tencent%29%20%28KDD%29%20Scenario-Adaptive%20Feature%20Interaction%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2024年（阿里巴巴）（CIKM） * [MultiLoRA] MultiLoRA——面向多域推荐的多方向低秩适配](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28Alibaba%29%20%28CIKM%29%20%2A%20%5BMultiLoRA%5D%20MultiLoRA%20-%20Multi-Directional%20Low-Rank%20Adaptation%20for%20Multi-Domain%20Recommendation.pdf) \u003Cbr \u002F>\n* [2024年（阿里巴巴）（RecSys） * [MLoRA] MLoRA——用于点击率预估的多域低秩适应网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28Alibaba%29%20%28RecSys%29%20%2A%20%5BMLoRA%5D%20MLoRA%20-%20Multi-Domain%20Low-Rank%20Adaptive%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2024年（快手）（SIGIR） [M3oE] M3oE——多域多任务专家混合推荐框架](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28Kuaishou%29%20%28SIGIR%29%20%5BM3oE%5D%20M3oE%20-%20Multi-Domain%20Multi-Task%20Mixture-of-Experts%20Recommendation%20Framework.pdf) \u003Cbr \u002F>\n* [2024年（腾讯）（KDD） [LCN] 面向在线点击率预估的跨域终身序列建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28Tencent%29%20%28KDD%29%20%5BLCN%5D%20Cross-Domain%20LifeLong%20Sequential%20Modeling%20for%20Online%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2024年（WSDM） 探索基于适配器的迁移学习在推荐系统中的应用——实证研究与实践启示](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2024%20%28WSDM%29%20Exploring%20Adapter-based%20Transfer%20Learning%20for%20Recommender%20Systems%20-%20Empirical%20Studies%20and%20Practical%20Insights.pdf) \u003Cbr \u002F>\n* [2025年（快手）（KDD） [HoME] HoME——快手多任务学习中的多门控专家层级](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-domain-Multi-Scenario\u002F2025%20%28Kuaishou%29%20%28KDD%29%20%5BHoME%5D%20HoME%20-%20Hierarchy%20of%20Multi-Gate%20Experts%20for%20Multi-Task%20Learning%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n\n#### 多任务\n* [(2018) (ICML) GradNorm - 用于深度多任务网络中自适应损失平衡的梯度归一化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F%282018%29%20%28ICML%29%20GradNorm%20-%20Gradient%20Normalization%20for%20Adaptive%20Loss%20Balancing%20in%20Deep%20Multitask%20Networks.pdf) \u003Cbr \u002F>\n* [2014 (TASLP) [LHUC] 用于无监督声学模型自适应的隐藏单元贡献学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2014%20%28TASLP%29%20%5BLHUC%5D%20Learning%20Hidden%20Unit%20Contributions%20for%20Unsupervised%20Acoustic%20Model%20Adaptation.pdf) \u003Cbr \u002F>\n* [2017 (Google) (ICLR) [Sparsely-Gated MOE] 极大规模神经网络——稀疏门控专家混合层](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2017%20%28Google%29%20%28ICLR%29%20%5BSparsely-Gated%20MOE%5D%20Outrageously%20large%20neural%20networks%20-%20The%20sparsely-gated%20mixture-of-experts%20layer.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (KDD) [DUPN] 深度感知用户——从多个电商任务中学习通用用户表示](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2018%20%28Alibaba%29%20%28KDD%29%20%5BDUPN%5D%20Perceive%20Your%20Users%20in%20Depth%20-%20Learning%20Universal%20User%20Representations%20from%20Multiple%20E-commerce%20Tasks.pdf) \u003Cbr \u002F>\n* [2018 (Alibaba) (SIGIR) [ESMM] 全空间多任务模型——一种有效估计点击后转化率的方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2018%20%28Alibaba%29%20%28SIGIR%29%20%5BESMM%5D%20Entire%20Space%20Multi-Task%20Model%20-%20An%20Effective%20Approach%20for%20Estimating%20Post-Click%20Conversion%20Rate.pdf) \u003Cbr \u002F>\n* [2018 (CVPR) 利用不确定性权衡场景几何与语义的多任务学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2018%20%28CVPR%29%20Multi-Task%20Learning%20Using%20Uncertainty%20to%20Weigh%20Losses%20for%20Scene%20Geometry%20and%20Semantics.pdf) \u003Cbr \u002F>\n* [2018 (Google) (KDD) ** [MMoE] 基于多门控专家混合体的多任务学习中任务关系建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2018%20%28Google%29%20%28KDD%29%20%2A%2A%20%5BMMoE%5D%20Modeling%20task%20relationships%20in%20multi-task%20learning%20with%20multi-gate%20mixture-of-experts.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (CIKM) 基于多任务的在线促销销量预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28Alibaba%29%20%28CIKM%29%20Multi-task%20based%20Sales%20Predictions%20for%20Online%20Promotions.pdf) \u003Cbr \u002F>\n* [2019 (Alibaba) (Recys) 一种在电商推荐中进行多目标优化的帕累托高效算法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28Alibaba%29%20%28Recys%29%20A%20Pareto-Eficient%20Algorithm%20for%20Multiple%20Objective%20Optimization%20in%20E-Commerce%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019 (Google) (AAAI) SNR子网络路由：用于多任务学习中的灵活参数共享](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28Google%29%20%28AAAI%29%20SNR%20Sub-Network%20Routing%20for%20Flexible%20Parameter%20Sharing%20in%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2019 (Google) (Recsys) ** [Youtube多任务] 推荐下一个观看的视频——一个多任务排序系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28Google%29%20%28Recsys%29%20%2A%2A%20%5BYoutube%20Multi-task%5D%20Recommending%20what%20video%20to%20watch%20next%20-%20a%20multitask%20ranking%20system.pdf) \u003Cbr \u002F>\n* [2019 (NIPS) 帕累托多任务学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2019%20%28NIPS%29%20Pareto%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (SIGIR) [ESM2] 通过点击后行为分解进行全空间多任务建模，以预测转化率](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Alibaba%29%20%28SIGIR%29%20%5BESM2%5D%20Entire%20Space%20Multi-Task%20Modeling%20via%20Post-Click%20Behavior%20Decomposition%20for%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020 (Alibaba) (WWW) 大规模因果方法：利用多任务学习去偏点击后转化率估计](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Alibaba%29%20%28WWW%29%20Large-scale%20Causal%20Approaches%20to%20Debiasing%20Post-click%20Conversion%20Rate%20Estimation%20with%20Multi-task%20Learning.pdf) \u003Cbr \u002F>\n* [2020 (Amazon) (WWW) 使用随机标签聚合的产品搜索多目标排序优化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Amazon%29%20%28WWW%29%20Multi-Objective%20Ranking%20Optimization%20for%20Product%20Search%20Using%20Stochastic%20Label%20Aggregation.pdf) \u003Cbr \u002F>\n* [2020 (Google) (KDD) [MoSE] 针对用户活动流的多任务序列专家混合体](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Google%29%20%28KDD%29%20%5BMoSE%5D%20Multitask%20Mixture%20of%20Sequential%20Experts%20for%20User%20Activity%20Streams.pdf) \u003Cbr \u002F>\n* [2020 (JD) (CIKM) *[DMT] 用于大规模电商推荐系统中多目标排序的深层多面Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28JD%29%20%28CIKM%29%20%2A%5BDMT%5D%20Deep%20Multifaceted%20Transformers%20for%20Multi-objective%20Ranking%20in%20Large-Scale%20E-commerce%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2020 (腾讯) (Recsys) **  [PLE] 渐进式分层提取（PLE）——一种用于个性化推荐的新型多任务学习（MTL）模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2020%20%28Tencent%29%20%28Recsys%29%20%2A%2A%20%5BPLE%5D%20Progressive%20Layered%20Extraction%20%28PLE%29%20-%20A%20Novel%20Multi-Task%20Learning%20%28MTL%29%20Model%20for%20Personalized%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (SIGIR) [HM3] 通过多任务学习分层建模微观和宏观行为，以预测转化率](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Alibaba%29%20%28SIGIR%29%20%5BMHM3%5D%20Hierarchically%20Modeling%20Micro%20and%20Macro%20Behaviors%20via%20Multi-Task%20Learning%20for%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Alibaba) (SIGIR) [MSSM] MSSM——一种高效的多任务学习多层次稀疏共享模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Alibaba%29%20%28SIGIR%29%20%5BMSSM%5D%20MSSM%20-%20A%20Multiple-level%20Sparse%20Sharing%20Model%20for%20Efficient%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2021 (百度) (SIGIR) [GemNN] GemNN——带有特征交互学习的门控增强型多任务神经网络，用于CTR预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Baidu%29%20%28SIGIR%29%20%5BGemNN%5D%20GemNN%20-%20Gating-Enhanced%20Multi-Task%20Neural%20Networks%20with%20Feature%20Interaction%20Learning%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2021 (Google) (Arxiv) [DSelect-k] 专家混合体中的可微选择及其在多任务学习中的应用](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Google%29%20%28Arxiv%29%20%5BDSelect-k%5D%20DSelect-k%20Differentiable%20Selection%20in%20the%20Mixture%20of%20Experts%20with%20Applications%20to%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2021 (Google) (KDD) 理解并改进多任务学习中的公平性与准确性权衡](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Google%29%20%28KDD%29%20Understanding%20and%20Improving%20Fairness-Accuracy%20Trade-offs%20in%20Multi-Task%20Learning.pdf) \u003Cbr \u002F>\n* [2021 (JD) (ICDE) 带有类别层次软约束的对抗性专家混合体](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28JD%29%20%28ICDE%29%20Adversarial%20Mixture%20Of%20Experts%20with%20Category%20Hierarchy%20Soft%20Constraint.pdf) \u003Cbr \u002F>\n* [2021 (美团) (KDD) 在定向展示广告中，利用多任务学习建模受众多步转化之间的顺序依赖关系](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Meituan%29%20%28KDD%29%20Modeling%20the%20Sequential%20Dependence%20among%20Audience%20Multi-step%20Conversions%20with%20Multi-task%20Learning%20in%20Targeted%20Display%20Advertising.pdf) \u003Cbr \u002F>\n* [2021 (腾讯) (Arxiv) 用于多目标用户画像建模的虚拟内核专家混合体](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Tencent%29%20%28Arxiv%29%20Mixture%20of%20Virtual-Kernel%20Experts%20for%20Multi-Objective%20User%20Profile%20Modeling.pdf) \u003Cbr \u002F>\n* [2021 (腾讯) (WWW) 个性化近似帕累托有效的推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2021%20%28Tencent%29%20%28WWW%29%20Personalized%20Approximate%20Pareto-Efficient%20Recommendation.pdf) \u003Cbr \u002F>\n* [2022 (Google) (WWW) 小头部能有所帮助吗？理解并改进多任务泛化能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2022%20%28Google%29%20%28WWW%29%20Can%20Small%20Heads%20Help%3F%20Understanding%20and%20Improving%20Multi-Task%20Generalization.pdf) \u003Cbr \u002F>\n* [2023 (Airbnb) (KDD) 利用多任务学习优化Airbnb搜索旅程](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Airbnb%29%20%28KDD%29%20Optimizing%20Airbnb%20Search%20Journey%20with%20Multi-task%20Learning.pdf) \u003Cbr \u002F>\n* [2023 (Alibaba) (CIKM) [DTRN] 用于多任务推荐的深度任务特定底层表示网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Alibaba%29%20%28CIKM%29%20%5BDTRN%5D%20Deep%20Task-specific%20Bottom%20Representation%20Network%20for%20Multi-Task%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023 (Google) (CIKM) 面向沉浸式信息流且无需再点击的多任务排序系统——短视频推荐案例研究](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Google%29%20%28CIKM%29%20Multitask%20Ranking%20System%20for%20Immersive%20Feed%20and%20No%20More%20Clicks%20-%20A%20Case%20Study%20of%20Short-Form%20Video%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023 (Google) (KDD) 改善推荐系统中多任务排序模型的训练稳定性](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Google%29%20%28KDD%29%20Improving%20Training%20Stability%20for%20Multitask%20Ranking%20Models%20in%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2023 (Meta) (KDD) AdaTT——用于推荐领域多任务学习的自适应任务间融合网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2023%20%28Meta%29%20%28KDD%29%20AdaTT%20-%20Adaptive%20Task-to-Task%20Fusion%20Network%20for%20Multitask%20Learning%20in%20Recommendations.pdf) \u003Cbr \u002F>\n* [2024 (Airbnb) (KDD) 通过模型蒸馏进行多目标排序学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Airbnb%29%20%28KDD%29%20Multi-objective%20Learning%20to%20Rank%20by%20Model%20Distillation.pdf) \u003Cbr \u002F>\n* [2024 (快手) (KDD) [GradCraft] GradCraft——通过整体梯度设计提升多任务推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Kuaishou%29%20%28KDD%29%20%5BGradCraft%5D%20GradCraft%20-%20Elevating%20Multi-task%20Recommendations%20through%20Holistic%20Gradient%20Crafting.pdf) \u003Cbr \u002F>\n* [2024 (快手) [HoME] HoME——快手多任务学习中的多门专家层级结构](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Kuaishou%29%20%5BHoME%5D%20HoME%20-%20Hierarchy%20of%20Multi-Gate%20Experts%20for%20Multi-Task%20Learning%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2024 (Shopee) (KDD) [ResFlow] 用于应用排序的残差多任务学习者](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Shopee%29%20%28KDD%29%20%5BResFlow%5D%20Residual%20Multi-Task%20Learner%20for%20Applied%20Ranking.pdf) \u003Cbr \u002F>\n* [2024 (腾讯) (KDD) [STEM] 广告推荐在崩塌且纠缠的世界中](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2024%20%28Tencent%29%20%28KDD%29%20%5BSTEM%5D%20Ads%20Recommendation%20in%20a%20Collapsed%20and%20Entangled%20World.pdf) \u003Cbr \u002F>\n* [2025 (百度) (KDD) [RankExpert] RankExpert——一种文本与行为专家混合体，用于网络搜索中的多目标排序学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2025%20%28Baidu%29%20%28KDD%29%20%5BRankExpert%5D%20RankExpert%20-%20A%20Mixture%20of%20Textual-and-Behavioral%20Experts%20for%20Multi-Objective%20Learning-to-Rank%20in%20Web%20Search.pdf) \u003Cbr \u002F>\n* [2025 (字节跳动) (CIKM) [PMTA] PMTA——感知驱动的多任务Transformer网络，用于个性化的多领域适配](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2025%20%28Byetedance%29%20%28CIKM%29%20%5BPMTA%5D%20PMTA%20-%20Perception-Aware%20Multi-Task%20Transformer%20Network%20for%20Personalized%20Multi-Domain%20Adaptation.pdf) \u003Cbr \u002F>\n* [2025 （阿里巴巴) (CIKM) [MAL] 看得更远——多归因学习带来更好的转化率预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FMulti-task\u002F2025%20%EF%BC%88Alibaba%29%20%28CIKM%29%20%5BMAL%5D%20See%20Beyond%20a%20Single%20View%20-%20Multi-Attribution%20Learning%20Leads%20to%20Better%20Conversion%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n\n#### 参数服务器\n* [2014年（百度）（OSDI）利用参数服务器扩展分布式机器学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FParameterServer\u002F2014%20%28Baidu%29%20%28OSDI%29%20Scaling%20Distributed%20Machine%20Learning%20with%20the%20Parameter%20Server.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（DLP-KDD）[XDL] XDL——面向高维稀疏数据的工业级深度学习框架](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FParameterServer\u002F2019%20%28Alibaba%29%20%28DLP-KDD%29%20%5BXDL%5D%20XDL%20-%20An%20Industrial%20Deep%20Learning%20Framework%20for%20High-dimensional%20Sparse%20Data.pdf) \u003Cbr \u002F>\n* [2020年（字节跳动）（OSDI）[BytePS] 一种用于加速异构GPU-CPU集群中分布式DNN训练的统一架构](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FParameterServer\u002F2020%20%28Bytedance%29%20%28OSDI%29%20%5BBytePS%5D%20A%20Unified%20Architecture%20for%20Accelerating%20Distributed%20DNN%20Training%20in%20Heterogeneous%20GPU%3ACPU%20Clusters.pdf) \u003Cbr \u002F>\n* [2022年（快手）（KDD）[Persia] Persia——一个开放的混合系统，可将基于深度学习的推荐系统扩展至100万亿参数](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FParameterServer\u002F2022%20%28Kuaishou%29%20%28KDD%29%20%5BPersia%5D%20Persia%20-%20An%20Open%2C%20Hybrid%20System%20Scaling%20Deep%20Learning-based%20Recommenders%20up%20to%20100%20Trillion%20Parameters.pdf) \u003Cbr \u002F>\n\n#### 预训练\n* [2019年（阿里巴巴）（IJCAI）[DeepMCP] 基于表示学习的点击率预估](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FPre-training\u002F2019%20%28Alibaba%29%20%28IJCAI%29%20%5BDeepMCP%5D%20Representation%20Learning-Assisted%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（SIGIR）[BERT4Rec] （阿里巴巴）（SIGIR2019）BERT4Rec——基于Transformer双向编码器表示的序列化推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FPre-training\u002F2019%20%28SIGIR%29%20%5BBERT4Rec%5D%20%28Alibaba%29%20%28SIGIR2019%29%20BERT4Rec%20-%20Sequential%20Recommendation%20with%20Bidirectional%20Encoder%20Representations%20from%20Transformer.pdf) \u003Cbr \u002F>\n\n#### 序列建模\n* [2016年（谷歌）（RecSys）**[YouTube DNN] 用于YouTube推荐的深度神经网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2016%20%28Google%29%20%28RecSys%29%20%2A%2A%5BYoutube%20DNN%5D%20Deep%20Neural%20Networks%20for%20YouTube%20Recommendations.pdf) \u003Cbr \u002F>\n* [2017年（谷歌）（NIPS）**注意力就是一切](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2017%20%28Google%29%20%28NIPS%29%20%2A%2A%20Attention%20Is%20All%20You%20Need.pdf) \u003Cbr \u002F>\n* [2018年（阿里巴巴）（KDD）**[DIN] 用于点击率预估的深度兴趣网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2018%20%28Alibaba%29%20%28KDD%29%20%2A%2A%5BDIN%5D%20Deep%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2018年（阿里巴巴）（KDD）[DUPN] 深度感知用户——从多个电商任务中学习通用用户表示](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2018%20%28Alibaba%29%20%28KDD%29%20%5BDUPN%5D%20Perceive%20Your%20Users%20in%20Depth%20-%20Learning%20Universal%20User%20Representations%20from%20Multiple%20E-commerce%20Tasks.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（AAAI）**[DIEN] 用于点击率预估的深度兴趣演化网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28AAAI%29%20%2A%2A%5BDIEN%5D%20Deep%20Interest%20Evolution%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（IJCAI）[DSIN] 用于点击率预估的深度会话兴趣网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28IJCAI%29%20%5BDSIN%5D%20Deep%20Session%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（KDD）[BST] 阿里巴巴电商推荐中的行为序列Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BBST%5D%20Behavior%20Sequence%20Transformer%20for%20E-commerce%20Recommendation%20in%20Alibaba.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（KDD）[DSTN] 用于点击率预估的深度时空神经网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BDSTN%5D%20Deep%20Spatio-Temporal%20Neural%20Networks%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（WWW）[TiSSA] TiSSA——一种基于时间切片自注意力的序列用户行为建模方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Alibaba%29%20%28WWW%29%20%5BTiSSA%5D%20TiSSA%20-%20A%20Time%20Slice%20Self-Attention%20Approach%20for%20Modeling%20Sequential%20User%20Behaviors.pdf) \u003Cbr \u002F>\n* [2019年（腾讯）（KDD）[RALM] 面向推荐系统的实时注意力相似模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2019%20%28Tencent%29%20%28KDD%29%20%5BRALM%5D%20TReal-time%20Attention%20Based%20Look-alike%20Model%20for%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（SIGIR）[DHAN] 用于点击率预估的层次化注意力深度兴趣网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28Alibaba%29%20%28SIGIR%29%20%5BDHAN%5D%20Deep%20Interest%20with%20Hierarchical%20Attention%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020年（谷歌）（KDD）[Google Drive] 提升Google Drive中的推荐质量](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28Google%29%20%28KDD%29%20%5BGoogle%20Drive%5D%20Improving%20Recommendation%20Quality%20in%20Google%20Drive.pdf) \u003Cbr \u002F>\n* [2020年（京东）（CIKM）**[DMT] 用于大规模电商推荐系统多目标排序的深度多面Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28JD%29%20%28CIKM%29%20%2A%2A%5BDMT%5D%20Deep%20Multifaceted%20Transformers%20for%20Multi-objective%20Ranking%20in%20Large-Scale%20E-commerce%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2020年（京东）（NIPS）[KFAtt] 用于CTR预估中用户行为建模的卡尔曼滤波注意力机制](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28JD%29%20%28NIPS%29%20%5BKFAtt%5D%20Kalman%20Filtering%20Attention%20for%20User%20Behavior%20Modeling%20in%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2020年（京东）（WSDM）[HUP] 用于电商推荐系统的层次化用户画像](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2020%20%28JD%29%20%28WSDM%29%20%5BHUP%5D%20Hierarchical%20User%20Profiling%20for%20E-commerce%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（WSDM）[RACP] 用于电商搜索中点击率预估的用户情境化页面级反馈建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2022%20%28Alibaba%29%20%28WSDM%29%20%5BRACP%5D%20Modeling%20Users%E2%80%99%20Contextualized%20Page-wise%20Feedback%20for%20Click-Through%20Rate%20Prediction%20in%20E-commerce%20Search.pdf) \u003Cbr \u002F>\n* [2022年（京东）（WWW）通过候选商品进行隐式用户感知建模以用于搜索广告的CTR预估](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2022%20%28JD%29%20%28WWW%29%20Implicit%20User%20Awareness%20Modeling%20via%20Candidate%20Items%20for%20CTR%20Prediction%20in%20Search%20Ads.pdf) \u003Cbr \u002F>\n* [2022年（WWW）[FMLP] 对于序列推荐，只需增强型MLP即可](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2022%20%28WWW%29%20%5BFMLP%5D%20Filter-enhanced%20MLP%20is%20All%20You%20Need%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2023年（京东）（CIKM）[IUI] IUI——用于点击率预估的意图增强型用户兴趣建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2023%20%28JD%29%20%28CIKM%29%20%5BIUI%5D%20IUI%20-%20Intent-Enhanced%20User%20Interest%20Modeling%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023年（美团）（CIKM）[DCIN] 用于点击率预估的深度上下文兴趣网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2023%20%28Meituan%29%20%28CIKM%29%20%5BDCIN%5D%20Deep%20Context%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2023年（Pinterest）（KDD）TransAct——基于Transformer的Pinterest实时用户行为推荐模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2023%20%28Pinterest%29%20%28KDD%29%20TransAct%20-%20Transformer-based%20Realtime%20User%20Action%20Model%20for%20Recommendation%20at%20Pinterest.pdf) \u003Cbr \u002F>\n* [2025年（快手）（SIGIR）[FIM] FIM——面向本地生活服务推荐的频率感知多视角兴趣建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling\u002F2025%20%28Kuaishou%29%20%28SIGIR%29%20%5BFIM%5D%20FIM%20-%20Frequency-Aware%20Multi-View%20Interest%20Modeling%20for%20Local-Life%20Service%20Recommendation.pdf) \u003Cbr \u002F>\n\n#### 序列建模-长期\n* [2019年（阿里巴巴）（KDD）[MIMN] 用于点击率预估的长序列用户行为建模实践](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BMIMN%5D%20Practice%20on%20Long%20Sequential%20User%20Behavior%20Modeling%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（WWW）面向长时序依赖用户序列的神经混合推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2019%20%28Google%29%20%28WWW%29%20Towards%20Neural%20Mixture%20Recommender%20for%20Long%20Range%20Dependent%20User%20Sequences.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（Arxiv）** [SIM] 基于搜索的用户兴趣建模：利用终身序列行为数据进行点击率预估](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2020%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BSIM%5D%20Search-based%20User%20Interest%20Modeling%20with%20Lifelong%20Sequential%20Behavior%20Data%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020年（ICLR）Reformer - 高效的Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2020%20%28ICLR%29%20Reformer%20-%20The%20Efficient%20Transformer%20.pdf) \u003Cbr \u002F>\n* [2020年（SIGIR）[UBR4CTR] 用于点击率预估的用户行为检索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2020%20%28SIGIR%29%20%5BUBR4CTR%5D%20User%20Behavior%20Retrieval%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（Arxiv）[ETA] 点击率预估模型中的端到端用户行为检索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2021%20%28Alibaba%29%20%28Arxiv%29%20%5BETA%5D%20End-to-End%20User%20Behavior%20Retrieval%20in%20Click-Through%20Rate%20Prediction%20Model.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（Arxiv）** [ETA] 用于点击率预估的高效长序列用户数据建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2022%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BETA%5D%20Efficient%20Long%20Sequential%20User%20Data%20Modeling%20for%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2022年（美团）（CIKM）[SDIM] 对于CTR预估中长期用户行为建模，采样就够了](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2022%20%28Meituan%29%20%28CIKM%29%20%5BSDIM%5D%20Sampling%20Is%20All%20You%20Need%20on%20Modeling%20Long-Term%20User%20Behaviors%20for%20CTR%20Prediction.pdf) \u003Cbr \u002F>\n* [2023年（快手）（Arixiv）[TWIN] TWIN - 双阶段兴趣网络：用于快手CTR预估中的终身用户行为建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2023%20%28Kuaishou%29%20%28Arixiv%29%20%5BTWIN%5D%20TWIN%20-%20TWo-stage%20Interest%20Network%20for%20Lifelong%20User%20Behavior%20Modeling%20in%20CTR%20Prediction%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2023年（快手）（CIKM）[QIN] 面向大规模搜索排序的查询主导型用户兴趣网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2023%20%28Kuaishou%29%20%28CIKM%29%20%5BQIN%5D%20Query-dominant%20User%20Interest%20Network%20for%20Large-Scale%20Search%20Ranking.pdf) \u003Cbr \u002F>\n* [2024年（快手）（CIKM）[TWINv2] TWIN V2 - 扩展超长用户行为序列建模，以提升快手的点击率预估能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2024%20%28Kuaishou%29%20%28CIKM%29%20%5BTWINv2%5D%20TWIN%20V2%20-%20Scaling%20Ultra-Long%20User%20Behavior%20Sequence%20Modeling%20for%20Enhanced%20CTR%20Prediction%20at%20Kuaishou.pdf) \u003Cbr \u002F>\n* [2024年（腾讯）（KDD）[LCN] 面向在线点击率预估的跨领域终身序列建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2024%20%28Tencent%29%20%28KDD%29%20%5BLCN%5D%20Cross-Domain%20LifeLong%20Sequential%20Modeling%20for%20Online%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（Arxiv）[MUSE] MUSE - 一个简单而高效的多模态搜索框架，用于终身用户兴趣建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BMUSE%5D%20MUSE%20-%20A%20Simple%20Yet%20Effective%20Multimodal%20Search-Based%20Framework%20for%20Lifelong%20User%20Interest%20Modeling.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）**（Arxiv）[LONGER] LONGER - 扩大工业级推荐系统中的长序列建模规模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BLONGER%5D%20LONGER%20-%20Scaling%20Up%20Long%20Sequence%20Modeling%20in%20Industrial%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025年（快手）（KDD）[HiT-LBM] 基于层次树搜索的用户终身行为建模，应用于大型语言模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Kuaishou%29%20%28KDD%29%20%5BHiT-LBM%5D%20Hierarchical%20Tree%20Search-based%20User%20Lifelong%20Behavior%20Modeling%20on%20Large%20Language%20Model.pdf) \u003Cbr \u002F>\n* [2025年（Meta）（Arxiv）[VISTA] 具有数百万亿参数的大规模记忆功能，适用于序列转换器生成式推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Meta%29%20%28Arxiv%29%20%5BVISTA%5D%20Massive%20Memorization%20with%20Hundreds%20of%20Trillions%20of%20Parameters%20for%20Sequential%20Transducer%20Generative%20Recommenders.pdf) \u003Cbr \u002F>\n* [2025年（Meta）（KDD）DV365 - 在Instagram上进行极长用户历史建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Meta%29%20%28KDD%29%20DV365%20-%20Extremely%20Long%20User%20History%20Modeling%20at%20Instagram.pdf) \u003Cbr \u002F>\n* [2025年（Pinterest）（Arxiv）[TransActV2] TransAct V2 - Pinterest推荐系统中的终身用户行为序列建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%28Pinterest%29%20%28Arxiv%29%20%5BTransActV2%5DTransAct%20V2%20-%20Lifelong%20User%20Action%20Sequence%20Modeling%20on%20Pinterest%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025年（字节跳动）**（Arxiv）[STCA] 要做就做长、要快就保持快 - 在抖音上实现百亿规模下的端到端1万序列建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FSequence-Modeling-Longterm\u002F2025%20%EF%BC%88Bytedance%29%20%2A%2A%20%28Arxiv%29%20%5BSTCA%5D%20Make%20It%20Long%2C%20Keep%20It%20Fast%20-%20End-to-End%2010k-Sequence%20Modeling%20at%20Billion%20Scale%20on%20Douyin.pdf) \u003Cbr \u002F>\n\n#### 迁移学习\n* [2014年（谷歌）（NIPS）[知识蒸馏] 蒸馏神经网络中的知识](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002F2014%20%28Google%29%20%28NIPS%29%20%5BKnoledge%20Distillation%5D%20Distilling%20the%20Knowledge%20in%20a%20Neural%20Network.pdf) \u003Cbr \u002F>\n* [2015年（ICLR）[Fitnets] Fitnets - 精简深度网络的提示](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002F2015%20%28ICLR%29%20%5BFitnets%5D%20Fitnets%20-%20Hints%20for%20thin%20deep%20nets.pdf) \u003Cbr \u002F>\n* [2018年（阿里巴巴）（AAAI）[Rocket] 火箭发射 - 一种通用且高效的框架，用于训练性能优异的轻量级网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002F2018%20%28Alibaba%29%20%28AAAI%29%20%5BRocket%5D%20Rocket%20launching%20-%20A%20universal%20and%20efficient%20framework%20for%20training%20well-performing%20light%20net.pdf) \u003Cbr \u002F>\n* [2018年（KDD）[排序蒸馏] 排序蒸馏 - 学习适用于推荐系统的高性能紧凑型排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002F2018%20%28KDD%29%5BRanking%20Distillation%5D%20Ranking%20distillation%20-%20Learning%20compact%20ranking%20models%20with%20high%20performance%20for%20recommender%20system.pdf) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002FCross-domain) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTransfer_Learning\u002FTransfer) \u003Cbr \u002F>\n\n#### 触发机制\n* [2022年（阿里巴巴）（WWW）触发式推荐中的点击率预估深度兴趣突出网络](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F03_Ranking\u002FTrigger\u002F2022%20%28Alibaba%29%20%28WWW%29%20Deep%20Interest%20Highlight%20Network%20for%20Click-Through%20Rate%20Prediction%20in%20Trigger-Induced%20Recommendation.pdf) \u003Cbr \u002F>\n\n## 04_排序后处理\n* [1998年（SIGIR）**  [MRR] 基于多样性的MMR重排序在文档重新排序与摘要生成中的应用](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F1998%20%28SIGIR%29%20%2A%2A%20%20%5BMRR%5D%20The%20Use%20of%20MMR%2C%20Diversity-Based%20Reranking%20for%20Reordering%20Documents%20and%20Producing%20Summaries.pdf) \u003Cbr \u002F>\n* [2005年（WWW）通过主题多样化改进推荐列表](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2005%20%28WWW%29%20Improving%20Recommendation%20Lists%20Through%20Topic%20Diversification.pdf) \u003Cbr \u002F>\n* [2008年（SIGIR）[α-NDCG] 信息检索评价中的新颖性与多样性](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2008%20%28SIGIR%29%20%5B%CE%B1-NDCG%5D%20Novelty%20and%20Diversity%20in%20Information%20Retrieval%20Evaluation.pdf) \u003Cbr \u002F>\n* [2009年（微软）（WSDM）搜索结果的多样化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2009%20%28Microsoft%29%20%28WSDM%29%20Diversifying%20Search%20Results.pdf) \u003Cbr \u002F>\n* [2010年（WWW）利用查询改写实现网页搜索结果的多样化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2010%20%28WWW%29%20Exploiting%20Query%20Reformulations%20for%20Web%20Search%20Result%20Diversification.pdf) \u003Cbr \u002F>\n* [2016年（亚马逊）（RecSys）面向视觉发现的自适应个性化多样性](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2016%20%28Amazon%29%20%28RecSys%29%20Adaptive%2C%20Personalized%20Diversity%20for%20Visual%20Discovery.pdf) \u003Cbr \u002F>\n* [2017年（Hulu）（NIPS）[DPP] 快速贪婪MAP推断用于行列式点过程以提升推荐多样性](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2017%20%28Hulu%29%20%28NIPS%29%20%5BDPP%5D%20Fast%20Greedy%20MAP%20Inference%20for%20Determinantal%20Point%20Process%20to%20Improve%20Recommendation%20Diversity.pdf) \u003Cbr \u002F>\n* [2018年（阿里巴巴）（IJCAI）[阿里巴巴GMV] 电商搜索中全局优化的相互影响感知排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2018%20%28Alibaba%29%20%28IJCAI%29%20%5BAlibaba%20GMV%5D%20Globally%20Optimized%20Mutual%20Influence%20Aware%20Ranking%20in%20E-Commerce%20Search.pdf) \u003Cbr \u002F>\n* [2018年（谷歌）（CIKM）[DPP] 使用行列式点过程在YouTube上实现实用的多样化推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2018%20%28Google%29%20%28CIKM%29%20%5BDPP%5D%20Practical%20Diversified%20Recommendations%20on%20YouTube%20with%20Determinantal%20Point%20Processes.pdf) \u003Cbr \u002F>\n* [2018年（SIGIR）[DLCM] 学习深度列表上下文模型以优化排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2018%20%28SIGIR%29%20%5BDLCM%5D%20Learning%20a%20Deep%20Listwise%20Context%20Model%20for%20Ranking%20Refinement.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（WWW）[基于价值的强化学习] 基于强化学习利润最大化的价值感知推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Alibaba%29%20%28WWW%29%20%5BValue-based%20RL%5D%20Value-aware%20Recommendation%20based%20on%20Reinforcement%20Profit%20Maximization.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（KDD）[GAttN] 通过最大团优化实现Exact-K推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Alibaba%29%20%28KDD%29%20%5BGAttN%5D%20Exact-K%20Recommendation%20via%20Maximal%20Clique%20Optimization.pdf) \u003Cbr \u002F>\n* [2019年（阿里巴巴）（RecSys）** [PRM] 推荐系统的个性化重排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Alibaba%29%20%28RecSys%29%20%2A%2A%20%5BPRM%5D%20Personalized%20Re-ranking%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（Arxiv）基于Slate的推荐系统强化学习——一种可处理的分解及实用方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Google%29%20%28Arxiv%29%20Reinforcement%20Learning%20for%20Slate-based%20Recommender%20Systems%20-%20A%20Tractable%20Decomposition%20and%20Practical%20Methodology.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（Arxiv）Seq2slate——使用RNN进行重排序与排版优化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Google%29%20%28Arxiv%29%20Seq2slate%20-%20Re-ranking%20and%20slate%20optimization%20with%20rnns.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（IJCAI）[SlateQ] SLATEQ——一种针对推荐集合强化学习的可处理分解](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2019%20%28Google%29%20%28IJCAI%29%20%5BSlateQ%5D%20SLATEQ%20-%20A%20Tractable%20Decomposition%20for%20Reinforcement%20Learning%20with%20Recommendation%20Sets.pdf) \u003Cbr \u002F>\n* [2020年（Airbnb）（KDD）管理Airbnb搜索中的多样性](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2020%20%28Airbnb%29%20%28KDD%29%20Managing%20Diversity%20in%20Airbnb%20Search.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（CIKM）[EdgeRec] EdgeRec——手机淘宝端边缘计算推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2020%20%28Alibaba%29%20%28CIKM%29%20%5BEdgeRec%5D%20EdgeRec%20-%20Recommender%20System%20on%20Edge%20in%20Mobile%20Taobao.pdf) \u003Cbr \u002F>\n* [2020年（华为）（Arxiv）个性化重排序以提升实时推荐系统的多样性](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2020%20%28Huawei%29%20%28Arxiv%29%20Personalized%20Re-ranking%20for%20Improving%20Diversity%20in%20Live%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（Arxiv）**  [PRS] 从排列视角重审推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2021%20%28Alibaba%29%20%28Arxiv%29%20%2A%2A%20%5BPRS%5D%20Revisit%20Recommender%20System%20in%20the%20Permutation%20Prospective.pdf) \u003Cbr \u002F>\n* [2021年（谷歌）（WSDM）用户响应模型以改进REINFORCE推荐系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2021%20%28Google%29%20%28WSDM%29%20User%20Response%20Models%20to%20Improve%20a%20REINFORCE%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2021年（微软）随时随地的多样性！在最大诱导基数目标下的流式行列式点过程](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2021%20%28Microsoft%29%20Diversity%20on%20the%20Go%21%20Streaming%20Determinantal%20Point%20Processes%20under%20a%20Maximum%20Induced%20Cardinality%20Objective.pdf) \u003Cbr \u002F>\n* [2023年（亚马逊）（KDD）RankFormer——使用列表级标签的列表式学习排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2023%20%28Amazon%29%20%28KDD%29%20RankFormer%20-%20Listwise%20Learning-to-Rank%20Using%20Listwide%20Labels.pdf) \u003Cbr \u002F>\n* [2023年（美团）（KDD）PIER——基于排列级别的兴趣端到端电商重排序框架](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2023%20%28Meituan%29%20%28KDD%29%20PIER%20-%20Permutation-Level%20Interest-Based%20End-to-End%20Re-ranking%20Framework%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2024年（快手）（KDD）[NAR4Rec] 非自回归生成模型用于推荐重排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2024%20%28Kuaishou%29%20%28KDD%29%20%5BNAR4Rec%5D%20Non-autoregressive%20Generative%20Models%20for%20Reranking%20Recommendation.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（SIGIR）[SORT-Gen] 针对淘宝列表级多目标优化的生成式重排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002F2025%20%28Alibaba%29%20%28SIGIR%29%20%5BSORT-Gen%5D%20A%20Generative%20Re-ranking%20Model%20for%20List-level%20Multi-objective%20Optimization%20at%20Taobao.pdf) \u003Cbr \u002F>\n\n#### Seq2Slate\n* [2015年（Google）（Arxiv） 大型离散动作空间中的深度强化学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2015%20%28Google%29%20%28Arxiv%29%20%20Deep%20Reinforcement%20Learning%20in%20Large%20Discrete%20Action%20Spaces%20.pdf) \u003Cbr \u002F>\n* [2015年（Google）（Arxiv） 带注意力机制的深度强化学习在高维状态与动作的Slate马尔可夫决策过程中的应用](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2015%20%28Google%29%20%28Arxiv%29%20Deep%20Reinforcement%20Learning%20with%20Attention%20for%20Slate%20Markov%20Decision%20Processes%20with%20High-Dimensional%20States%20and%20Actions.pdf) \u003Cbr \u002F>\n* [2017年（KDD）[DCM] 使用指针网络的深度选择模型用于航空公司行程预测](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2017%20%28KDD%29%20%5BDCM%5D%20Deep%20Choice%20Model%20Using%20Pointer%20Networks%20for%20Airline%20Itinerary%20Prediction.pdf) \u003Cbr \u002F>\n* [2018年（Microsoft）（EMNLP）[RL4NMT] 针对神经机器翻译的强化学习研究](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2018%20%28Microsoft%29%20%28EMNLP%29%20%5BRL4NMT%5D%20A%20study%20of%20reinforcement%20learning%20for%20neural%20machine%20translation.pdf) \u003Cbr \u002F>\n* [2019年（Google）（Arxiv）Seq2slate——利用RNN进行重排序与展示列表优化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F04_Post-ranking\u002FSeq2Slate\u002F2019%20%28Google%29%20%28Arxiv%29%20Seq2slate%20-%20Re-ranking%20and%20slate%20optimization%20with%20rnns.pdf) \u003Cbr \u002F>\n\n## 05_相关性排序\n* [2013年（微软）（CIKM）[DSSM] 利用点击数据学习用于网络搜索的深度结构化语义模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2013%20%28Microsoft%29%20%28CIKM%29%20%5BDSSM%5D%20Learning%20deep%20structured%20semantic%20models%20for%20web%20search%20using%20clickthrough%20data.pdf) \u003Cbr \u002F>\n* [2016年（雅虎）（KDD）雅虎搜索中的相关性排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2016%20%28Yahoo%29%20%28KDD%29%20Ranking%20Relevance%20in%20Yahoo%20Search.pdf) \u003Cbr \u002F>\n* [2020年（ICLR）[StructBERT] StructBERT - 将语言结构融入预训练以实现深度语言理解](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2020%20%28ICLR%29%20%5BStructBERT%5D%20StructBERT%20-%20Incorporating%20Language%20Structures%20into%20Pre-training%20for%20Deep%20Language%20Understanding.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（WWW）[MASM] 从电商点击数据中学习商品相关性模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2021%20%28Alibaba%29%20%28WWW%29%20%5BMASM%5D%20Learning%20a%20Product%20Relevance%20Model%20from%20Click-Through%20Data%20in%20E-Commerce.pdf) \u003Cbr \u002F>\n* [2022年（阿里巴巴）（KDD）[ReprBERT] ReprBERT - 将BERT蒸馏为高效的基于表示的相关性模型，用于电子商务](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2022%20%28Alibaba%29%20%28KDD%29%20%5BReprBERT%5D%20ReprBERT%20-%20Distilling%20BERT%20to%20an%20Efficient%20Representation-Based%20Relevance%20Model%20for%20E-Commerce.pdf) \u003Cbr \u002F>\n* [2023年（美团）（CIKM）[SPM] SPM - 面向美团搜索相关性建模的结构化预训练与匹配架构](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2023%20%28Meituan%29%20%28CIKM%29%20%5BSPM%5D%20SPM%20-%20Structured%20Pretraining%20and%20Matching%20Architectures%20for%20Relevance%20Modeling%20in%20Meituan%20Search.pdf) \u003Cbr \u002F>\n* [2024年（阿里巴巴）（KDD）[DeepBoW] 深度词袋模型 - 一种高效且可解释的中文电商相关性架构](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2024%20%28Alibaba%29%20%28KDD%29%20%5BDeepBoW%5D%20Deep%20Bag-of-Words%20Model%20-%20An%20Efficient%20and%20Interpretable%20Relevance%20Architecture%20for%20Chinese%20E-Commerce.pdf) \u003Cbr \u002F>\n* [2024年（沃尔玛）（SIGIR）用于商品搜索相关性判断的大语言模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2024%20%EF%BC%88Walmart%29%20%28SIGIR%29%20Large%20Language%20Models%20for%20Relevance%20Judgment%20in%20Product%20Search.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（Arxiv）LORE - 用于搜索相关性的大型生成模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20LORE%20-%20A%20Large%20Generative%20Model%20for%20Search%20Relevance.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（WWW）[ELLM] 可解释的LLM驱动多维蒸馏在电商相关性学习中的应用](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%28Alibaba%29%20%28WWW%29%20%5BELLM%5D%20Explainable%20LLM-driven%20Multi-dimensional%20Distillation%20for%20E-Commerce%20Relevance%20Learning.pdf) \u003Cbr \u002F>\n* [2025年（快手）（Arxiv）[HCMRM] HCMRM - 用于搜索广告的高一致性多模态相关性模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%28Kuaishou%29%20%28Arxiv%29%20%5BHCMRM%5D%20HCMRM%20-A%20High-Consistency%20Multimodal%20Relevance%20Model%20for%20Search%20Ads.pdf) \u003Cbr \u002F>\n* [2025年（LinkedIn）（Arxiv）大规模求职搜索的动力 - 职位匹配系统中增强型LLM查询理解](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%28Linkedin%29%20%28Arxiv%29%20Powering%20Job%20Search%20at%20Scale%20-%20LLM-Enhanced%20Query%20Understanding%20in%20Job%20Matching%20Systems.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（Arxiv）TaoSR1 - 电商相关性搜索的思维模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%EF%BC%88Alibaba%29%20%28Arxiv%29%20TaoSR1%20-%20The%20Thinking%20Model%20for%20E-commerce%20Relevance%20Search.pdf) \u003Cbr \u002F>\n* [2025年（腾讯）（KDD）[GenFR] 在腾讯中应用大语言模型进行相关性搜索](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F05_Relevance-ranking\u002F2025%20%EF%BC%88Tencent%29%20%28KDD%29%20%5BGenFR%5D%20Applying%20Large%20Language%20Model%20For%20Relevance%20Search%20In%20Tencent.pdf) \u003Cbr \u002F>\n\n## 06_LLM\n\n#### 01_LLM_经典\n* [2013年（谷歌）（NIPS）[Word2vec] 单词和短语的分布式表示及其组合性](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2013%20%28Google%29%20%28NIPS%29%20%5BWord2vec%5D%20Distributed%20Representations%20of%20Words%20and%20Phrases%20and%20their%20Compositionality.pdf) \u003Cbr \u002F>\n* [2014年（谷歌）（NIPS）[Seq2Seq] 基于神经网络的序列到序列学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2014%20%28Google%29%20%28NIPS%29%20%5BSeq2Seq%5D%20Sequence%20to%20Sequence%20Learning%20with%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2017年（谷歌）（NIPS）[Transformer] “注意力就是你所需要的”](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2017%20%28Google%29%20%28NIPS%29%20%5BTransformer%5D%20Attention%20Is%20All%20You%20Need.pdf) \u003Cbr \u002F>\n* [2017年（OpenAI）（NIPS）[RLHF] 基于人类偏好的深度强化学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2017%20%28OpenAI%29%20%28NIPS%29%20%5BRLHF%5D%20Deep%20Reinforcement%20Learning%20from%20Human%20Preferences.pdf) \u003Cbr \u002F>\n* [2018年（OpenAI）（Arxiv）[GPT-1] 通过生成式预训练提升语言理解能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2018%20%28OpenAI%29%20%28Arxiv%29%20%5BGPT-1%5D%20Improving%20Language%20Understanding%20by%20Generative%20Pre-Training.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（NAACL）[Bert] BERT——面向语言理解的深度双向Transformer预训练](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2019%20%28Google%29%20%28NAACL%29%20%5BBert%5D%20BERT%20-%20Pre-training%20of%20Deep%20Bidirectional%20Transformers%20for%20Language%20Understanding.pdf) \u003Cbr \u002F>\n* [2019年（OpenAI）（Arxiv）[GPT-2] 语言模型是无监督的多任务学习者](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2019%20%28OpenAI%29%20%28Arxiv%29%20%5BGPT-2%5D%20Language%20Models%20are%20Unsupervised%20Multitask%20Learners.pdf) \u003Cbr \u002F>\n* [2020年（Arxiv）神经语言模型的规模法则](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2020%20%28Arxiv%29%20Scaling%20Laws%20for%20Neural%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2020年（Meta）（NIPS）[RAG] 面向知识密集型NLP任务的检索增强生成](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2020%20%28Meta%29%20%28NIPS%29%20%5BRAG%5D%20Retrieval-Augmented%20Generation%20for%20Knowledge-Intensive%20NLP%20Tasks.pdf) \u003Cbr \u002F>\n* [2020年（OpenAI）（Arxiv）[GPT-3] 语言模型是少样本学习者](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2020%20%28OpenAI%29%20%28Arxiv%29%20%5BGPT-3%5D%20Language%20Models%20are%20Few-Shot%20Learners.pdf) \u003Cbr \u002F>\n* [2021年（微软）（Arxiv）[LoRA] LoRA——大型语言模型的低秩适应](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2021%20%28Microsoft%29%20%28Arxiv%29%20%5BLoRA%5D%20LoRA%20-%20Low-Rank%20Adaptation%20of%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2022年（谷歌）（Arxiv）[PaLM] PaLM——通过Pathways扩展语言建模](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28Arxiv%29%20%5BPaLM%5D%20PaLM%20-%20Scaling%20Language%20Modeling%20with%20Pathways.pdf) \u003Cbr \u002F>\n* [2022年（谷歌）（JMLR）[SwitchTransfomers] Switch Transformers——以简单高效的稀疏性扩展至万亿参数模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28JMLR%29%20%5BSwitchTransfomers%5D%20Switch%20Transformers%20-%20Scaling%20to%20Trillion%20Parameter%20Models%20with%20Simple%20and%20Efficient%20Sparsity.pdf) \u003Cbr \u002F>\n* [2022年（谷歌）（NIPS）[COT] 思维链提示在大型语言模型中激发推理能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28NIPS%29%20%5BCOT%5D%20Chain-of-Thought%20Prompting%20Elicits%20Reasoning%20in%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2022年（谷歌）（NIPS）[ChainOfThought] 思维链提示在大型语言模型中激发推理能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28NIPS%29%20%5BChainOfThought%5D%20Chain-of-Thought%20Prompting%20Elicits%20Reasoning%20in%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2022年（谷歌）（TMLR）[Emergent] 大型语言模型的涌现能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28Google%29%20%28TMLR%29%20%5BEmergent%5D%20Emergent%20Abilities%20of%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2022年（OpenAI）（Arxiv）[InstructGPT] [RLHF] 通过人类反馈训练语言模型遵循指令](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28OpenAI%29%20%28Arxiv%29%20%5BInstructGPT%5D%20%5BRLHF%5D%20Training%20language%20models%20to%20follow%20instructions%20with%20human%20feedback.pdf) \u003Cbr \u002F>\n* [2022年（OpenAI）（Arxiv）[PPO] 近端策略优化算法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28OpenAI%29%20%28Arxiv%29%20%5BPPO%5D%20Proximal%20Policy%20Optimization%20Algorithms.pdf) \u003Cbr \u002F>\n* [2022年（OpenAI）（Arxiv）[WebGPT] 从人类反馈中学习总结](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28OpenAI%29%20%28Arxiv%29%20%5BWebGPT%5D%20Learning%20to%20summarize%20from%20human%20feedback.pdf) \u003Cbr \u002F>\n* [2022年（OpenAI）（Arxiv）[WebGPT] WebGPT——基于浏览器的人工协助问答系统](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2022%20%28OpenAI%29%20%28Arxiv%29%20%5BWebGPT%5D%20WebGPT%20-%20Browser-assisted%20question-answering%20with%20human%20feedback.pdf) \u003Cbr \u002F>\n* [2023年（阿里巴巴）（Arxiv）[QWEN] QWEN技术报告](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2023%20%28Alibaba%29%20%28Arxiv%29%20%5BQWEN%5D%20QWEN%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2023年（Meta）（Arxiv）[LLaMA-2] Llama 2——开放的基础模型与微调后的聊天模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2023%20%28Meta%29%20%28Arxiv%29%20%5BLLaMA-2%5D%20Llama%202%20-%20Open%20Foundation%20and%20Fine-Tuned%20ChatModels.pdf) \u003Cbr \u002F>\n* [2023年（Meta）（Arxiv）[LLaMA] LLaMA——开放且高效的语言基础模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2023%20%28Meta%29%20%28Arxiv%29%20%5BLLaMA%5D%20LLaMA%20-%20Open%20and%20Efficient%20Foundation%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2023年（OpenAI）（Arxiv）[GPT4] GPT-4技术报告](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2023%20%28OpenAI%29%20%28Arxiv%29%20%5BGPT4%5D%20GPT-4%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2024年（阿里巴巴）（Arxiv）[QWEN2] QWEN2技术报告](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2024%20%28Alibaba%29%20%28Arxiv%29%20%5BQWEN2%5D%20QWEN2%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2024年（Arxiv）[TinyLlama] Arxiv TinyLlama——一个开源的小型语言模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2024%20%28Arxiv%29%20%5BTinyLlama%5D%20Arxiv%20TinyLlama%20-%20An%20Open-Source%20Small%20Language%20Model.pdf) \u003Cbr \u002F>\n* [2024年（DeepSeek）（Arxiv）[GRPO] DeepSeekMath——推动开放语言模型中的数学推理极限](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2024%20%28DeepSeek%29%20%28Arxiv%29%20%5BGRPO%5D%20DeepSeekMath%20-%20Pushing%20the%20Limits%20of%20Mathematical%20Reasoning%20in%20Open%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（Arxiv）[QWEN-2.5] QWEN 2.5技术报告](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BQWEN-2.5%5D%20QWEN%202.5%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（Arxiv）[Qwen2.5-VL] Qwen2.5-VL技术报告](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BQwen2.5-VL%5D%20Qwen2.5-VL%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（Arxiv）[Qwen3 Embedding] Qwen3嵌入——通过基础模型推进文本嵌入与重排序](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BQwen3%20Embedding%5D%20Qwen3%20Embedding%20-%20Advancing%20Text%20Embedding%20and%20Reranking%20Through%20Foundation%20Models.pdf) \u003Cbr \u002F>\n* [2025年（阿里巴巴）（Arxiv）[Qwen3] Qwen3技术报告](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Alibaba%29%20%28Arxiv%29%20%5BQwen3%5D%20Qwen3%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025年（Arxiv）大型语言模型综述](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Arxiv%29%20A%20Survey%20of%20Large%20Language%20Models.pdf) \u003Cbr \u002F>\n* [2025年（DeepSeek）（Nature）[DeepSeek-R1] DeepSeek-R1通过强化学习激励LLM中的推理能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28DeepSeek%29%20%28Nature%29%20%5BDeepSeek-R1%5D%20DeepSeek-R1%20incentivizes%20reasoning%20in%20LLMs%20through%20reinforcement%20learning.pdf) \u003Cbr \u002F>\n* [2025年（DeepSeek）[DeepSeek-R1] DeepSeek-R1——通过强化学习激励LLM中的推理能力](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28DeepSeek%29%20%5BDeepSeek-R1%5D%20DeepSeek-R1%20-Incentivizing%20Reasoning%20Capability%20in%20LLMs%20via%20Reinforcement%20Learning.pdf) \u003Cbr \u002F>\n* [2025年（DeepSeek）[DeepSeek-V3] DeepSeek-V3技术报告](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28DeepSeek%29%20%5BDeepSeek-V3%5D%20DeepSeek-V3%20Technical%20Report.pdf) \u003Cbr \u002F>\n* [2025年（谷歌）（Arxiv）[SigLIP2] SigLIP 2——具有改进语义理解、定位能力和密集特征的多语言视觉-语言编码器](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002F2025%20%28Google%29%20%28Arxiv%29%20%5BSigLIP2%5D%20SigLIP%202%20-%20Multilingual%20Vision-Language%20Encoders%20with%20Improved%20Semantic%20Understanding%2C%20Localization%2C%20and%20Dense%20Features.pdf) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FBook) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FMOE) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FModelOptimization) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FMultiModal) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FQuant) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FSFT) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FSpecificApplication) \u003Cbr \u002F>\n* [](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002FSurvey) \u003Cbr \u002F>\n* [资源](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F01_LLM_Classical\u002Fresources.txt) \u003Cbr \u002F>\n\n#### 02_自监督学习\n* [2020年（阿里巴巴）（AAAI）[DMR] 用于个性化点击率预估的深度匹配排序模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Alibaba%29%20%28AAAI%29%20%5BDMR%5D%20Deep%20Match%20to%20Rank%20Model%20for%20Personalized%20Click-Through%20Rate%20Prediction.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（CIKM）[BERT4Rec] BERT4Rec - 基于Transformer双向编码器表示的序列推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Alibaba%29%20%28CIKM%29%20%5BBERT4Rec%5D%20BERT4Rec%20-%20Sequential%20Recommendation%20with%20Bidirectional%20Encoder%20Representations%20from%20Transformer.pdf) \u003Cbr \u002F>\n* [2020年（阿里巴巴）（KDD）序列推荐中的解耦自监督学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Alibaba%29%20%28KDD%29%20Disentangled%20Self-Supervision%20in%20Sequential%20Recommenders.pdf) \u003Cbr \u002F>\n* [2020年（Arxiv）UserBERT - 自监督用户表征学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Arxiv%29%20UserBERT%20-%20Self-supervised%20User%20Representation%20Learning.pdf) \u003Cbr \u002F>\n* [2020年（Arxiv）[SGL] 用于推荐的自监督图学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28Arxiv%29%20%5BSGL%5D%20Self-supervised%20Graph%20Learning%20for%20Recommendation.pdf) \u003Cbr \u002F>\n* [2020年（CIKM）[S3Rec] S3-Rec - 基于互信息最大化的序列推荐自监督学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28CIKM%29%20%5BS3Rec%5D%20S3-Rec%20-%20Self-Supervised%20Learning%20for%20Sequential%20Recommendation%20with%20Mutual%20Information%20Maximization.pdf) \u003Cbr \u002F>\n* [2020年（EMNLP）[PTUM] PTUM - 基于自监督从无标签用户行为中预训练用户模型](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28EMNLP%29%20%5BPTUM%5D%20PTUM%20-%20Pre-training%20User%20Model%20from%20Unlabeled%20User%20Behaviors%20via%20Self-supervision.pdf) \u003Cbr \u002F>\n* [2020年（SIGIR）推荐系统的自监督强化学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2020%20%28SIGIR%29%20Self-Supervised%20Reinforcement%20Learning%20for%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（Arxiv）[CLRec] 大规模推荐系统中去偏候选生成的对比学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28Alibaba%29%20%28Arxiv%29%20%5BCLRec%5D%20Contrastive%20Learning%20for%20Debiased%20Candidate%20Generation%20in%20Large-Scale%20Recommender%20Systems.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（CIKM）* [ZEUS] 针对电商多场景排序的用户自发行为自监督学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28Alibaba%29%20%28CIKM%29%20%2A%20%5BZEUS%5D%20Self-Supervised%20Learning%20on%20Users%E2%80%99%20Spontaneous%20Behaviors%20for%20Multi-Scenario%20Ranking%20in%20E-commerce.pdf) \u003Cbr \u002F>\n* [2021年（阿里巴巴）（WWW）序列推荐的对比预训练](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28Alibaba%29%20%28WWW%29%20Contrastive%20Pre-training%20for%20Sequential%20Recommendation.pdf) \u003Cbr \u002F>\n* [2021年（Google）（CIKM）大规模物品推荐的自监督学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28Google%29%20%28CIKM%29%20Self-supervised%20Learning%20for%20Large-scale%20Item%20Recommendations.pdf) \u003Cbr \u002F>\n* [2021年（WSDM）[Prop] PROP - 面向即席检索的代表性词预测预训练](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F06_LLM\u002F02_Self_Supervised_Learning\u002F2021%20%28WSDM%29%20%5BProp%5D%20PROP%20-%20Pre-training%20with%20Representative%20Words%20Prediction%20for%20Ad-hoc%20Retrieval.pdf) \u003Cbr \u002F>\n\n## 07_强化学习\n* [2010年（雅虎）（WWW）[LinUCB] 基于上下文多臂老虎机的个性化新闻文章推荐方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2010%20%28Yahoo%29%20%28WWW%29%20%5BLinUCB%5D%20A%20Contextual-Bandit%20Approach%20to%20Personalized%20News%20Article%20Recommendation.pdf) \u003Cbr \u002F>\n* [2018年（Spotify）（Recsys）[Spotify Bandit] 探索、利用与解释：用多臂老虎机实现可解释的个性化推荐](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2018%20%28Spotify%29%20%28Recsys%29%20%5BSpotify%20Bandit%5D%20Explore%2C%20Exploit%2C%20and%20Explain%20Personalizing%20Explainable%20Recommendations%20with%20Bandits.pdf) \u003Cbr \u002F>\n* [2018年[微软]（WWW）[DRN] DRN——用于新闻推荐的深度强化学习框架](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2018%20%5BMicrosoft%5D%20%28WWW%29%20%5BDRN%5D%20DRN%20-%20A%20Deep%20Reinforcement%20Learning%20Framework%20for%20News%20Recommendation.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（IJCAI）*[SlateQ] SLATEQ——面向推荐集合的强化学习中的一种可处理分解方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2019%20%28Google%29%20%28IJCAI%29%20%2A%5BSlateQ%5D%20SLATEQ%20-%20A%20Tractable%20Decomposition%20for%20Reinforcement%20Learning%20with%20Recommendation%20Sets.pdf) \u003Cbr \u002F>\n* [2019年（谷歌）（WSDM）*[Top-K离策略] REINFORCE推荐系统中的Top-K离策略校正](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2019%20%28Google%29%20%28WSDM%29%20%2A%5BTop-K%20Off-Policy%5D%20Top-K%20Off-Policy%20Correction%20for%20a%20REINFORCE%20Recommender%20System.pdf) \u003Cbr \u002F>\n* [2019年（Sigweb）搜索、推荐与在线广告领域的深度强化学习——综述](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2019%20%28Sigweb%29%20Deep%20Reinforcement%20Learning%20for%20Search%2C%20Recommendation%2C%20and%20Online%20Advertising%20-%20A%20Survey.pdf) \u003Cbr \u002F>\n* [2020年（字节跳动）（KDD）[RAM] 联合学习推荐与广告投放](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2020%20%28Bytedance%29%20%28KDD%29%20%5BRAM%5D%20Jointly%20Learning%20to%20Recommend%20and%20Advertise.pdf) \u003Cbr \u002F>\n* [2020年（京东）（SIGIR）[NICF] 神经交互式协同过滤](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002F2020%20%28JD%29%20%28SIGIR%29%20%5BNICF%5D%20Neural%20Interactive%20Collaborative%20Filtering.pdf) \u003Cbr \u002F>\n\n#### RL_classical\n* [1992年（ML）[REINFORCE] 连接主义强化学习中的简单统计梯度跟踪算法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F1992%20%28ML%29%20%5BREINFORCE%5D%20Simple%20statistical%20gradient-following%20algorithms%20for%20connectionist%20reinforcement%20learning.pdf) \u003Cbr \u002F>\n* [1999年（NIPS）[Actor-Critic] Actor-Critic算法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F1999%20%28NIPS%29%20%5BActor-Critic%5D%20Actor-Critic%20Algorithms.pdf) \u003Cbr \u002F>\n* [2013年（DeepMind）（Arxiv）[DQN] 使用深度强化学习玩Atari游戏](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2013%20%28DeepMind%29%20%28Arxiv%29%20%5BDQN%5D%20Playing%20Atari%20with%20Deep%20Reinforcement%20Learning.pdf) \u003Cbr \u002F>\n* [2015年（DeepMind）（AAAI）[Double Q-learning] 使用双Q学习的深度强化学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2015%20%28DeepMind%29%20%28AAAI%29%20%5BDouble%20Q-learning%5D%20Deep%20Reinforcement%20Learning%20with%20Double%20Q-learning.pdf) \u003Cbr \u002F>\n* [2015年（DeepMind）（Nature）[DQN] 通过深度强化学习实现人类水平的控制](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2015%20%28DeepMind%29%20%28Nature%29%20%5BDQN%5D%20Human-level%20control%20through%20deep%20reinforcement%20learning.pdf) \u003Cbr \u002F>\n* [2016年（谷歌）（Arxiv）[A3C] 深度强化学习的异步方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2016%20%28Google%29%20%28Arxiv%29%20%5BA3C%5D%20Asynchronous%20Methods%20for%20Deep%20Reinforcement%20Learning.pdf) \u003Cbr \u002F>\n* [2016年（OpenAI）（Nature）使用深度神经网络和树搜索掌握围棋游戏](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2016%20%28OpenAI%29%20%28Nature%29%20Mastering%20the%20game%20of%20Go%20with%20deep%20neural%20networks%20and%20tree%20search.pdf) \u003Cbr \u002F>\n* [2017年（JMLR）[TRPO] 信任域策略优化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2017%20%28JMLR%29%20%5BTRPO%5D%20Trust%20Region%20Policy%20Optimization.pdf) \u003Cbr \u002F>\n* [2017年（OpenAI）（Arxiv）[PPO] 近端策略优化算法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2017%20%28OpenAI%29%20%28Arxiv%29%20%5BPPO%5D%20Proximal%20Policy%20Optimization%20Algorithms.pdf) \u003Cbr \u002F>\n* [2017年（OpenAI）（Arxiv）[ES] 进化策略作为强化学习的可扩展替代方案](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F07_Reinforcement_Learning\u002FRL_classical\u002F2017%20%EF%BC%88OpenAI%29%20%28Arxiv%29%20%5BES%5D%20Evolution%20Strategies%20as%20a%20Scalable%20Alternative%20to%20Reinforcement%20Learning.pdf) \u003Cbr \u002F>\n\n## 08_深度学习\n* [2012年（NIPS）[CNN] 使用深度卷积神经网络进行ImageNet分类](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2012%20%28NIPS%29%20%5BCNN%5D%20ImageNet%20Classification%20with%20Deep%20Convolutional%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2014年（JMLR）[Dropout] Dropout——防止神经网络过拟合的简单方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2014%20%28JMLR%29%20%5BDropout%5D%20Dropout%20-%20A%20Simple%20Way%20to%20Prevent%20Neural%20Networks%20from%20Overfitting.pdf) \u003Cbr \u002F>\n* [2015年（Google）（JMLR）[BatchNorm] 批量归一化——通过减少内部协变量偏移加速深度网络训练](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2015%20%28Google%29%20%28JMLR%29%20%5BBatchNorm%5D%20Batch%20Normalization%20-%20Accelerating%20Deep%20Network%20Training%20by%20Reducing%20Internal%20Covariate%20Shift.pdf) \u003Cbr \u002F>\n* [2015年（OpenAI）（ICLR）[Adam] Adam——一种随机优化方法](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2015%20%28OpenAI%29%20%28ICLR%29%20%5BAdam%5D%20Adam%20-%20A%20Method%20for%20Stochastic%20Optimization.pdf) \u003Cbr \u002F>\n* [2016年（CVPR）[ResNet] 用于图像识别的深度残差学习](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2016%20%28CVPR%29%20%5BResNet%5D%20Deep%20Residual%20Learning%20for%20Image%20Recognition.pdf) \u003Cbr \u002F>\n* [2016年（OpenAI）（NIPS）[Weight Norm] 权重归一化——一种简单的重新参数化方法，可加速深度神经网络的训练](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2016%20%28OpenAI%29%20%28NIPS%29%20%5BWeight%20Norm%5D%20Weight%20Normalization%20-%20A%20Simple%20Reparameterization%20to%20Accelerate%20Training%20of%20Deep%20Neural%20Networks.pdf) \u003Cbr \u002F>\n* [2017年（Arxiv）[LayerNorm] 层归一化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2017%20%28Arxiv%29%20%5BLayerNorm%5D%20Layer%20Normalization.pdf) \u003Cbr \u002F>\n* [2017年（Google）（NIPS）[Transformer] “注意力就是你所需要的”](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2017%20%28Google%29%20%28NIPS%29%20%5BTransformer%5D%20Attention%20Is%20All%20You%20Need.pdf) \u003Cbr \u002F>\n* [2020年（Arxiv）GLU变体改进Transformer](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2020%20%28Arxiv%29%20GLU%20Variants%20Improve%20Transformer.pdf) \u003Cbr \u002F>\n* [2020年（ICML）关于Transformer架构中的层归一化](https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements\u002Fblob\u002Fmaster\u002F08_Deep_Learning\u002F2020%20%28ICML%29%20On%20Layer%20Normalization%20in%20the%20Transformer%20Architecture.pdf) \u003Cbr \u002F>","# Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising 快速上手指南\n\n本仓库并非一个可执行的软件工具或 Python 包，而是一个**精选论文清单**。它整理了工业界搜索、推荐和广告领域（涵盖 Embedding、匹配、排序、重排序等阶段）的经典深度学习论文。\n\n因此，本指南将指导你如何**获取资源**、**阅读核心文献**以及**复现经典模型**。\n\n## 环境准备\n\n由于本仓库主要提供论文 PDF 链接，阅读本身无需特殊环境。但若要复现论文中的算法，建议准备以下开发环境：\n\n*   **操作系统**: Linux (Ubuntu\u002FCentOS) 或 macOS (Windows 用户建议使用 WSL2)\n*   **编程语言**: Python 3.8+\n*   **深度学习框架**: PyTorch 或 TensorFlow (根据具体论文实现选择)\n*   **依赖管理**: `pip` 或 `conda`\n*   **网络环境**: 部分论文链接托管在 GitHub 上，国内访问可能较慢，建议配置代理或使用加速工具。\n\n## 安装步骤（获取资源）\n\n你可以通过克隆仓库的方式将所有论文索引和目录下载到本地。\n\n### 1. 克隆仓库\n打开终端，执行以下命令：\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements.git\n```\n\n> **国内加速方案**: 如果直接克隆速度过慢，可使用国内镜像源（如 Gitee 镜像，若存在）或指定代理：\n> ```bash\n> git clone https:\u002F\u002Fghproxy.com\u002Fhttps:\u002F\u002Fgithub.com\u002Fguyulongcs\u002FDeep-Learning-for-Search-Recommendation-Advertisements.git\n> ```\n> *(注：ghproxy.com 为常用的 GitHub 加速服务，若失效请切换回官方地址或使用科学上网)*\n\n### 2. 进入目录\n```bash\ncd Deep-Learning-for-Search-Recommendation-Advertisements\n```\n\n此时，你可以在本地文件夹中看到按技术阶段分类的子目录（如 `00_Embedding`, `01_Matching` 等），里面包含了指向具体论文 PDF 的 Markdown 链接文件。\n\n## 基本使用\n\n本项目的核心用法是**按需查阅论文**并**复现模型**。\n\n### 1. 查阅论文列表\n你可以直接在 GitHub 网页版浏览，或在本地用 Markdown 编辑器打开对应的 `.md` 文件（如果仓库包含汇总文件），直接点击链接下载 PDF。\n\n**核心分类导航：**\n*   **00_Embedding**: 包含 Word2vec, DeepWalk, Node2vec, GCN, GraphSAGE, PinSage 等图嵌入与表示学习论文。\n*   **01_Matching**: 包含 User-CF, Item-CF, DSSM, Youtube DNN, Two-Tower, MIND 等召回与匹配模型论文。\n*   *(其他分类如 Ranking, Re-ranking 等在完整仓库中继续列出)*\n\n### 2. 复现经典模型示例 (以 Youtube DNN 为例)\n假设你想复现 `01_Matching` 章节中的 **[Youtube DNN]** 模型，通常步骤如下：\n\n#### 第一步：阅读论文\n点击仓库中对应的链接下载并阅读 *Deep Neural Networks for YouTube Recommendations (2016)*。\n\n#### 第二步：寻找开源实现\n该仓库仅提供论文链接。你需要在 GitHub 搜索该论文的开源实现代码。例如：\n\n```bash\n# 在 GitHub 搜索 \"Youtube DNN pytorch implementation\"\n# 找到合适的仓库后克隆，例如 (以下为示例虚拟地址，实际请搜索最新高星项目):\ngit clone https:\u002F\u002Fgithub.com\u002Fexample-user\u002Fyoutube-dnn-pytorch.git\ncd youtube-dnn-pytorch\n```\n\n#### 第三步：安装依赖并运行\n进入代码目录，安装依赖并运行训练脚本：\n\n```bash\n# 创建虚拟环境\npython -m venv venv\nsource venv\u002Fbin\u002Factivate  # Windows 用户使用: venv\\Scripts\\activate\n\n# 安装依赖\npip install torch torchvision torchaudio --index-url https:\u002F\u002Fdownload.pytorch.org\u002Fwhl\u002Fcu118\npip install -r requirements.txt\n\n# 运行训练示例 (具体命令视下载的代码库而定)\npython train.py --config configs\u002Fyoutube_dnn.yaml\n```\n\n### 3. 构建自己的知识库\n你可以利用仓库中的分类结构，将自己复现代码的笔记或整理好的中文解读添加到对应目录下，构建个人的搜广推知识体系。\n\n---\n**提示**: 仓库中带 `*` 或 `**` 标记的论文通常为工业界落地效果显著的经典之作（如 Alibaba Embedding, MIND, Two-Tower），建议优先研读。","某电商平台的推荐算法团队正面临用户点击率增长瓶颈，急需引入工业界验证过的深度学习模型来优化商品嵌入（Embedding）与匹配策略。\n\n### 没有 Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising 时\n- **文献检索如大海捞针**：工程师需手动在 arXiv、Google Scholar 等平台搜索\"Graph Embedding\"或\"CTR Prediction\"，耗费数周才能拼凑出零散的论文列表，且难以区分学术理论与工业落地方案。\n- **技术选型缺乏依据**：面对 Word2vec、Node2vec、GCN 等众多模型，团队不清楚哪些是谷歌、阿里、Pinterest 等大厂在实际亿级数据场景中验证有效的，容易盲目尝试不成熟的算法。\n- **知识体系碎片化**：团队成员对从预排序到重排序的全链路技术认知不一，缺乏统一的参考标准，导致技术方案讨论时经常因信息不对称而陷入低效争论。\n\n### 使用 Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising 后\n- **一站式获取工业级方案**：团队直接查阅该清单中\"00_Embedding\"章节，迅速定位到阿里巴巴的 Billion-scale Commodity Embedding 和 Pinterest 的 PinSage 等经典论文，将调研周期从数周缩短至两天。\n- **精准对标落地场景**：通过清单中明确标注的厂商（如 Google、Alibaba）和应用场景（如电商推荐、社交网络），团队果断放弃纯学术模型，优先复现已在大规模生产中验证过的 GraphSAGE 和 GAT 算法。\n- **构建系统化技术图谱**：依托清单涵盖的嵌入、匹配、排序及强化学习等全链路分类，团队快速统一了技术视野，基于成熟的工业界演进路径制定了清晰的模型迭代路线图。\n\nAwesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising 将分散的顶会论文转化为结构化的工业实战指南，帮助算法团队大幅降低试错成本并加速核心指标提升。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fguyulongcs_Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising_12f69d10.png","guyulongcs","Yulong Gu","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fguyulongcs_c3ce0b49.jpg","Research Scientist in Online Advertisement, Recommendation, Search, Deep Learning, LLM.","Tsinghua University(PhD)","Beijing,China","guyulongcs@gmail.com",null,"https:\u002F\u002Fguyulongcs.github.io","https:\u002F\u002Fgithub.com\u002Fguyulongcs",[84],{"name":85,"color":86,"percentage":87},"Python","#3572A5",100,2427,286,"2026-04-09T06:14:24",1,"",{"notes":94,"python":92,"dependencies":95},"该仓库是一个深度学习论文合集（Awesome List），主要包含搜索、推荐和广告领域的学术论文链接（PDF），并非可执行的软件代码库。因此，它没有特定的操作系统、GPU、内存、Python 版本或依赖库要求。用户只需具备阅读 PDF 文档的能力即可使用。若需复现论文中的算法，则需参考各篇论文具体的实现要求。",[],[15,97,14],"其他",[99,100,101,102,103,104,105,106],"deep-learning","recommender-system","search","advertising","ctr","cvr","reinforcement-learning","search-engine","2026-03-27T02:49:30.150509","2026-04-09T20:54:33.646335",[],[]]