[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-HKUSTDial--NL2SQL_Handbook":3,"tool-HKUSTDial--NL2SQL_Handbook":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",149489,2,"2026-04-10T11:32:46",[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。",108322,"2026-04-10T11:39:34",[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},6121,"gemini-cli","google-gemini\u002Fgemini-cli","gemini-cli 是一款由谷歌推出的开源 AI 命令行工具，它将强大的 Gemini 大模型能力直接集成到用户的终端环境中。对于习惯在命令行工作的开发者而言，它提供了一条从输入提示词到获取模型响应的最短路径，无需切换窗口即可享受智能辅助。\n\n这款工具主要解决了开发过程中频繁上下文切换的痛点，让用户能在熟悉的终端界面内直接完成代码理解、生成、调试以及自动化运维任务。无论是查询大型代码库、根据草图生成应用，还是执行复杂的 Git 操作，gemini-cli 都能通过自然语言指令高效处理。\n\n它特别适合广大软件工程师、DevOps 人员及技术研究人员使用。其核心亮点包括支持高达 100 万 token 的超长上下文窗口，具备出色的逻辑推理能力；内置 Google 搜索、文件操作及 Shell 命令执行等实用工具；更独特的是，它支持 MCP（模型上下文协议），允许用户灵活扩展自定义集成，连接如图像生成等外部能力。此外，个人谷歌账号即可享受免费的额度支持，且项目基于 Apache 2.0 协议完全开源，是提升终端工作效率的理想助手。",100752,"2026-04-10T01:20:03",[52,13,15,14],"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":66,"readme_en":67,"readme_zh":68,"quickstart_zh":69,"use_case_zh":70,"hero_image_url":71,"owner_login":72,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":76,"owner_location":76,"owner_email":77,"owner_twitter":76,"owner_website":78,"owner_url":79,"languages":80,"stars":85,"forks":86,"last_commit_at":87,"license":76,"difficulty_score":32,"env_os":88,"env_gpu":89,"env_ram":89,"env_deps":90,"category_tags":93,"github_topics":94,"view_count":32,"oss_zip_url":76,"oss_zip_packed_at":76,"status":17,"created_at":110,"updated_at":111,"faqs":112,"releases":113},6324,"HKUSTDial\u002FNL2SQL_Handbook","NL2SQL_Handbook","This is a continuously updated handbook for readers to easily track the latest Text-to-SQL techniques in the literature and provide practical guidance for researchers and practitioners. ","NL2SQL_Handbook 是一本持续更新的权威指南，旨在帮助读者轻松追踪自然语言转 SQL（Text-to-SQL）领域的最新技术进展。它基于多篇顶级学术会议（如 VLDB、TKDE）的综述论文构建，核心目标是解决大模型时代下，从业者难以选择合适技术方案以及研究人员难以把握未来方向的痛点。\n\n该资源不仅提供了全面的技术调研和深度论文解读，还涵盖了从数据合成、基准测试到多维度评估及错误分析的全生命周期指导。其独特亮点在于将 Text-to-SQL 面临的挑战划分为五个层级，清晰梳理了从早期模型到当前大模型阶段的技术演进脉络，并展望了未来五年的系统发展愿景。\n\nNL2SQL_Handbook 非常适合数据库领域的研究人员、AI 工程师以及需要落地自然语言查询功能的企业开发者使用。对于希望深入理解如何将用户自然语言高效转化为数据库查询语句，或正在寻找特定场景最佳实践的专业人士而言，这是一份兼具理论深度与实战价值的必备参考手册。"," \u003Ch1 align=\"center\">Text-to-SQL Handbook\u003C\u002Fh1>\n\n \u003Ch3 align=\"center\">NL2SQL Handbook\u003C\u002Fh3>\n \nThis is the official repository for **[TKDE'25] A Survey of Text-to-SQL in the Era of LLMs: Where are we, and where are we going?** and **[VLDB'24] The Dawn of Natural Language to SQL: Are We Fully Ready?**.\nFrom this repository, you can explore the [latest advancements](#-text-to-sql-survey--tutorial) in Text-to-SQL research (a.k.a NL2SQL). We provide a comprehensive survey, in-depth research papers, and benchmark evaluations. \n\n**\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2025-green\"> A Survey of Text-to-SQL in the Era of LLMs: Where are we, and where are we going?**\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.05109)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSlides-orange\">](.\u002Fslides\u002FNL2SQL_handbook.pdf)\n\n**\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> Natural Language to SQL: State of the Art and Open Problems**\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdbgroup.cs.tsinghua.edu.cn\u002Fligl\u002Fpapers\u002FVLDB25-NL2SQL.pdf)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSlides-orange\">](.\u002Fslides\u002FNL2SQL-VLDB2025.pdf)\n\n**\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> The Dawn of Natural Language to SQL: Are We Fully Ready?**\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol17\u002Fp3318-luo.pdf) \n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSlides-orange\">](.\u002Fslides\u002FNL2SQL360-VLDB2024.pdf)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-purple\">](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL360)\n\n📧If we missed any interesting work, [connect with us](#connect-with-us).\n\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002Friver.svg\"\u002F>\n\u003C\u002Fp>\n\n```bibtex\n@article{liu2025survey,\n  title={A Survey of Text-to-SQL in the Era of LLMs: Where are we, and where are we going?},\n  author={Liu, Xinyu and Shen, Shuyu and Li, Boyan and Ma, Peixian and Jiang, Runzhi and Zhang, Yuxin and Fan, Ju and Li, Guoliang and Tang, Nan and Luo, Yuyu},\n  journal={IEEE Transactions on Knowledge and Data Engineering},\n  year={2025},\n  publisher={IEEE}\n}\n```\n\n## 🧭 Text-to-SQL Introduction \nTranslating users' natural language queries (NL) into SQL queries can significantly reduce barriers to accessing relational databases and support various commercial applications. The performance of Text-to-SQL has been greatly improved with the emergence of language models (LMs). In this context, it is crucial to assess our current position, determine the Text-to-SQL solutions that should be adopted for specific scenarios by practitioners, and identify the research topics that researchers should explore next.\n\n\u003Cp align=\"center\">\n\u003Cimg width=\"600\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FHKUSTDial_NL2SQL_Handbook_readme_8b45a8d62150.jpg\"\u002F>\n\u003C\u002Fp>\n\n## 📈 Text-to-SQL Lifecycle\n\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002Fnl2sql_lifecycle.svg\"\u002F>\n\u003C\u002Fp>\n\n+ Model: Text-to-SQL translation techniques that tackle not only NL ambiguity and under-specification, but also properly map NL with database schema and instances;\n\n+ Data: From the collection of training data, data synthesis due to training data scarcity, to Text-to-SQL benchmarks;\n\n+ Evaluation: Evaluating Text-to-SQL methods from multiple angles using different metrics and granularities;\n\n+ Error Analysis: analyzing Text-to-SQL errors to find the root cause and guiding Text-to-SQL models to evolve.\n\n## 🤔 Where Are We?\nWe categorize the challenges of Text-to-SQL into five levels, each addressing specific hurdles. The first three levels cover challenges that have been or are currently being addressed, reflecting the progressive development of Text-to-SQL. The fourth level represents the challenges we aim to tackle in the LLMs stage, while the fifth level outlines our vision for Text-to-SQL system in the next five years. \n\nWe describe the evolution of Text-to-SQL solutions from the perspective of language models, categorizing it into four stages.\nFor each stage of Text-to-SQL, we analyze the changes in target users and the extent to which challenges are addressed.\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002FThe Evolution of NL2SQL Solutions from the Perspective of Language Models.svg\"\u002F>\n\u003C\u002Fp>\n\n\n## 🧩 Module-based Text-to-SQL Methods\nWe summarize the key modules of Text-to-SQL solutions\nutilizing the language model. \n+ **Pre-processing** serves as an enhancement to the model’s inputs in the Text-to-SQL parsing process. You can get more details from this chapter: [Pre-Processing](chapter\u002FPre_Processing.md)\n+ **Text-to-SQL translation methods** constitute the core of the Text-to-SQL solution, responsible for converting input natural language queries into SQL queries. You can get more details from this chapter: [Text-to-SQL Translation Methods](chapter\u002FTranslation_method.md)\n+ **Post-processing** is a crucial step to refine the generated SQL queries, ensuring they meet user expectations more accurately. You can get more details from this chapter: [Post-Processing](chapter\u002FPost_Processing.md)\n\u003Cp align=\"center\">\n\u003Cimg width=\"600\" src=\".\u002Fassets\u002FAn Overview of NL2SQL Method in the LLM Era.svg\"\u002F>\n\u003C\u002Fp>\n\n## 📚 Text-to-SQL Survey & Tutorial\n\n1. A Survey of Text-to-SQL in the Era of LLMs:\nWhere are we, and where are we going?\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.05109) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL_Handbook)\n1. Natural Language to SQL: State of the Art and Open Problems. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdbgroup.cs.tsinghua.edu.cn\u002Fligl\u002Fpapers\u002FVLDB25-NL2SQL.pdf)\n1. Next-generation database interfaces: A survey of LLM-based Text-to-SQL.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.08426)\n1. A Survey on Employing Large Language Models for Text-to-SQL Tasks.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCSUR'2024-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2407.15186)\n1. Large Language Model Enhanced Text-to-SQL Generation: A Survey.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.06011)\n1. From Natural Language to SQL: Review of LLM-based Text-to-SQL Systems.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2410.01066)\n1. Natural language interfaces for tabular data querying and visualization: A survey.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2024-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.17894)\n1. Natural Language Interfaces for Databases with Deep Learning.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2023-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.14778\u002F3611540.3611575)\n1. A survey on deep learning approaches for text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDBJ'2023-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1007\u002Fs00778-022-00776-8)\n1. A Survey on Text-to-SQL Parsing: Concepts, Methods, and Future Directions.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2022-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2208.13629)\n1. Recent Advances in Text-to-SQL: A Survey of What We Have and What We Expect.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLING'2022-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2022.coling-1.190\u002F)\n1. A Deep Dive into Deep Learning Approaches for Text-to-SQL Systems.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2021-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1145\u002F3448016.3457543)\n1. State of the Art and Open Challenges in Natural Language Interfaces to Data.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2020-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1145\u002F3318464.3383128)\n1. Natural language to SQL: Where are we today? \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2020-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol13\u002Fp1737-kim.pdf)\n\n## 📰 Text-to-SQL Paper List\n1. DeepEye-SQL: A Software-Engineering-Inspired Text-to-SQL Framework. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2026-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.17586)\n1. Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Search\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICML'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.17248) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Falpha-sql-hkust.github.io\u002F)\n1. NL2SQL-BUGs: A Benchmark for Detecting Semantic Errors in NL2SQL Translation.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGKDD'2025-B6FFBB\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.11984) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fnl2sql-bugs.github.io\u002F)\n1. EllieSQL: Cost-Efficient Text-to-SQL with Complexity-Aware Routing. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLM'2025-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.22402) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Felliesql.github.io\u002F)\n1. The Dawn of Natural Language to SQL: Are We Fully Ready?\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.01265) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL360)\n2. DIVER: A Robust Text-to-SQL System with Dynamic Interactive Value Linking and Evidence Reasoning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2026-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2602.12064)\n3. LogicCat: A Chain-of-Thought Text-to-SQL Benchmark for Complex Reasoning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2026-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fojs.aaai.org\u002Findex.php\u002FAAAI\u002Farticle\u002Fview\u002F40243)\n4. Schema on the Inside: A Two-Phase Fine-Tuning Method for High-Efficiency Text-to-SQL at Scale. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2026-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fojs.aaai.org\u002Findex.php\u002FAAAI\u002Farticle\u002Fview\u002F41446)\n5. Hexgen-Flow: Optimizing LLM Inference Request Scheduling for Agentic Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.05286) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRelaxed-System-Lab\u002FHexgen-Flow)\n6. SQLMorph: Query Mutation and Fine-Grained Metrics for Text-to-SQL Evaluation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Famine.io\u002Fpapers\u002F2026-icde-sqlmorph.pdf)  [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fdais-polymtl\u002Fsqlmorph)\n7. LEAF-SQL: Level-wise Exploration with Adaptive Fine-graining for Text-to-SQL Skeleton Prediction. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\">  [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FAtlamtiz\u002FLEAF-SQL)\n8. An Efficient and Effective Evaluator for Text2SQL Models on Unseen and Unlabeled Data. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\">\n9. Boosting Small Language Models for Text-to-SQL with Fine-Grained Execution Feedback and Cost-Efficient Rewards. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\">\n10. MM2SQL: A Benchmark and Method for Visually-Grounded SQL Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> \n11. Text2SQL-Flow: A Robust SQL-Aware Data Augmentation Framework for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.10192)\n12. Knapsack Optimization-based Schema Linking for LLM-based Text-to-SQL Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.12911)\n13. CYANSQL: Unlock the Power of NL2SQL via Clustering-based Test-Time Scaling. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\">\n14. Beyond Static Pipelines: Learning Dynamic Workflows for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.15077) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FSatissss\u002FSquRL)\n15. ReViSQL: Achieving Human-Level Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.20004) \n16. Think2SQL: Blueprinting Reward Density and Advantage Scaling for Effective Text-to-SQL Reasoning.  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">]([https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.15077](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2602.15564)) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fanonymous.4open.science\u002Fr\u002FThink2SQL-3B7F\u002FREADME.md)\n1. AgentSM: Semantic Memory for Agentic Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2601.15709)\n1. LLM-Based SQL Generation: Prompting, Self-Refinement, and Adaptive Weighted Majority Voting. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2601.17942)\n1. Pervasive Annotation Errors Break Text-to-SQL Benchmarks and Leaderboards.  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.08778) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fuiuc-kang-lab\u002Ftext_to_sql_benchmarks)\n1. OptiSQL: Executable SQL Generation from Optical Tokens. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2601.13695) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FjohnnyZeppelin\u002FOptiSQL) \n1. Memo-SQL: Structured Decomposition and Experience-Driven Self-Correction for Training-Free NL2SQL.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.arxiv.org\u002Fpdf\u002F2601.10011)\n1. Structure-Guided Large Language Models for Text-to-SQL Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICML'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Ficml.cc\u002Fvirtual\u002F2025\u002Fposter\u002F44477)\n1. Sphinteract: Resolving Ambiguities in NL2SQL Through User Interaction.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol18\u002Fp1145-zhao.pdf) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FZhaoFuheng\u002FSphinteract)\n1. OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scale. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">]([https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol18\u002Fp1145-zhao.pdf](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.02240)) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRUCKBReasoning\u002FOmniSQL)\n1. EVOSCHEMA: TOWARDS TEXT-TO-SQL ROBUSTNESS AGAINST SCHEMA EVOLUTION. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fpdf?id=NfUHBaZdLw) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fzhangtianshu\u002FEvoSchema)\n1. Is Long Context All You Need? Leveraging LLM's Extended Context for NL2SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.12372) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fyeounoh\u002Flc_nl2sql)\n1. The Power of Constraints in Natural Language to SQL Translation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol18\u002Fp2097-ren.pdf)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fhttdty\u002FREDSQL_VLDB)\n1. OpenSearch-SQL: Enhancing Text-to-SQL with Dynamic Few-shot and Consistency Alignment. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2025-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.14913) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FOpenSearch-AI\u002FOpenSearch-SQL)\n1. Reliable Text-to-SQL with Adaptive Abstention.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2025-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.10858) \n1. SNAILS: Schema Naming Assessments for Improved LLM-Based SQL Inference.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2025-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1145\u002F3709727)\n1. Automated Validating and Fixing of Text-to-SQL Translation with Execution Consistency. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2025-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fipads.se.sjtu.edu.cn\u002Fzh\u002Fpublications\u002FSQLDriller.pdf)\n1. Grounding Natural Language to SQL Translation with Data-Based Self-Explanations.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.02948) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FKaimary\u002FCycleSQL)\n1. AID-SQL: Adaptive In-Context Learning of Text-to-SQL with Difficulty-Aware Instruction and Retrieval-Augmented Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.computer.org\u002Fcsdl\u002Fproceedings-article\u002Ficde\u002F2025\u002F360300d945\u002F26FZCc99mg0) \n1. CLEAR: A Parser-Independent Disambiguation Framework for NL2SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.computer.org\u002Fcsdl\u002Fproceedings-article\u002Ficde\u002F2025\u002F360300d302\u002F26FZBD2hBJe) \n1. CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICLR'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2410.01943v1) \n1. Spider 2.0: Evaluating Language Models on Real-World Enterprise Text-to-SQL Workflows.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICLR'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.07763) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fxlang-ai\u002FSpider2)\n1. ROUTE: Robust Multitask Tuning and Collaboration for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICLR'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2412.10138)\n1. SHARE: An SLM-based Hierarchical Action CorREction Assistant for Text-to-SQL.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.00391)\n1. DCG-SQL: Enhancing In-Context Learning for Text-to-SQL with Deep Contextual Schema Link Graph. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.19956)\n1. Uncovering the Impact of Chain-of-Thought Reasoning for Direct Preference Optimization: Lessons from Text-to-SQL.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.11656)\n1. STaR-SQL: Self-Taught Reasoner for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13550)\n1. SQLGenie: A Practical LLM based System for Reliable and Efficient SQL Generation \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025(industry)-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">]([https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13550](https:\u002F\u002Faclanthology.org\u002F2025.acl-industry.71\u002F))\n1. SQL-R1: Training Natural Language to SQL Reasoning Model By Reinforcement Learning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNeurIPS'2025-yellow\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.08600) \n1. Confidence Estimation for Error Detection in Text-to-SQL Systems. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2025-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.09527)\n1. SQLord: A Robust Enterprise Text-to-SQL Solution via Reverse Data Generation and Workflow Decomposition. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWWW'2025-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fpdf\u002F10.1145\u002F3701716.3715541)\n1. DBCopilot: Scaling Natural Language Querying to Massive Databases.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEDBT\u002FICDT'2025-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.03463) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Ftshu-w\u002FDBCopilot)\n1. Utilising Large Language Models for Adversarial Attacks in Text-to-SQL: A Perpetrator and Victim Approach.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBTW'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.20657) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FXGenerationLab\u002FXiYan-DBDescGen)\n1. You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNAACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.12172) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fsig4kg.github.io\u002Farcher-bench\u002F)\n1. Boosting Text-to-SQL through Multi-grained Error Identification.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLING'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2025.coling-main.289.pdf)\n1. Gen-SQL: Efficient Text-to-SQL By Bridging Natural Language Question And Database Schema With Pseudo-Schema.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLING'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2025.coling-main.256\u002F)\n1. MAC-SQL: A Multi-Agent Collaborative Framework for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLING'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.11242) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fwbbeyourself\u002FMAC-SQL)\n1. PARSQL: Enhancing Text-to-SQL through SQL Parsing and Reasoning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2024.findings-acl.120\u002F)\n1. UCS-SQL: Uniting Content and Structure for Enhanced Semantic Bridging In Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fforum?id=xnTouV7wyr)\n1. SQLForge: Synthesizing Reliable and Diverse Data to Enhance Text-to-SQL Reasoning in LLMs. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.13725)\n1. Optimizing Reasoning for Text-to-SQL with Execution Feedback. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.19988)\n1. Knowledge Base Construction for Knowledge-Augmented Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2505.22096)\n1. SQLong: Enhanced NL2SQL for Longer Contexts with LLMs.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Workshop)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.16747)\n1. Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLM'2025-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.23157)\n1. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBIRD Top1-red\">Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scaling.  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.24403).\n1. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBIRD Top2-blue\">Automatic Metadata Extraction for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.19988)\n1. DeepEye-SQL: A Software-Engineering-Inspired Text-to-SQL Framework. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.17586)\n1. Rethinking Text-to-SOL: Dynamic Multi-turn SOIInteraction for Real-world Database Exploration. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2510.26495) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FAurora-slz\u002FRealWorld-SQL-Bench)\n1. MARS-SQL: A MULTI-AGENT REINFORCEMENT LEARNING FRAMEWORK FOR TEXT-TO-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2511.01008) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FYangHaolin0526\u002FMARS-SQL)\n1. RUBIKSQL: Lifelong Learning Agentic Knowledge Base as an Industrial NL2SQL System. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2508.17590) \n1. CSC-SQL: Corrective Self-Consistency in Text-to-SQL via Reinforcement Learning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.13271) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FCycloneBoy\u002Fcsc_sql\u002F)\n1. Cheaper, Better, Faster, Stronger: Robust Text-to-SQL without Chain-of-Thought or Fine-Tuning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2505.14174) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fgenaasia\u002FN-rep)\n1. SLM-SQL: An Exploration of Small Language Models for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.22478) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FCycloneBoy\u002Fslm_sql)\n1. Reward-SQL: Boosting Text-to-SQL via Stepwise Reasoning and Process-Supervised Rewards. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2505.04671) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fruc-datalab\u002FRewardSQL)\n1. Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.20315) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fsnowflakedb\u002FArcticTraining)\n1. Corrective Self-Consistency in Text-to-SQL via Reinforcement Learning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.13271) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FCycloneBoy\u002Fcsc_sql)\n1. SQLForge: Synthesizing Reliable and Diverse Data to Enhance Text-to-SQL Reasoning in LLMs. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2505.13725)\n1. Think2SQL: Reinforce LLM Reasoning Capabilities for Text2SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.15077)\n1. Distill-C: Enhanced NL2SQL via Distilled Customization with LLMs. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.00048)\n1. OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scale. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.02240) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRUCKBReasoning\u002FOmniSQL)\n1. SQL-Factory: A Multi-Agent Framework for High-Quality and Large-Scale SQL Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.14837)\n1. Text2SQL is Not Enough: Unifying AI and Databases with TAG. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2408.14717) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FTAG-Research\u002FTAG-Bench) \n1. Automatic database description generation for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.20657)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FXGenerationLab\u002FXiYan-DBDescGen)\n1. MCTS-SQL: An Effective Framework for Text-to-SQL with Monte Carlo Tree Search.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.16607)\n1. SQL-o1: A Self-Reward Heuristic Dynamic Search Method for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.11741)\n1. FEATHER-SQL: A Lightweight NL2SQL Framework with Dual-Model Collaboration Paradigm for Small Language Models.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.17811)\n1. FI-NL2PY2SQL: Financial Industry NL2SQL Innovation Model Based on Python and Large Language Model.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F17\u002F1\u002F12)\n1. FGCSQL: A Three-Stage Pipeline for Large Language Model-Driven Chinese Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.mdpi.com\u002F2079-9292\u002F14\u002F6\u002F1214)\n1. Transforming Medical Data Access: The Role and Challenges of Recent Language Models in SQL Query Automation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.mdpi.com\u002F1999-4893\u002F18\u002F3\u002F124)\n1. The Dawn of Natural Language to SQL: Are We Fully Ready?\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.01265) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL360)\n1. Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.15363) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FBeachWang\u002FDAIL-SQL) \n1. Interleaving Pre-Trained Language Models and Large Language Models for Zero-Shot NL2SQL Generation. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.08891) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fruc-datalab\u002FZeroNL2SQL)\n1. Generating Succinct Descriptions of Database Schemata for Cost-Efficient Prompting of Large Language Models. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.14778\u002F3681954.3682017) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fitrummer\u002Fschemacompression)\n1. ScienceBenchmark: A Complex Real-World Benchmark for Evaluating Natural Language to SQL Systems.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.04743) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fsciencebenchmark.cloudlab.zhaw.ch\u002F)\n1. CodeS: Towards Building Open-source Language Models for Text-to-SQL. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2024-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.16347) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRUCKBReasoning\u002Fcodes)\n1. FinSQL: Model-Agnostic LLMs-based Text-to-SQL Framework for Financial Analysis. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2024-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.10506) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fbigbigwatermalon\u002FFinSQL)\n1. PURPLE: Making a Large Language Model a Better SQL Writer. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2024-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.20014) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fhttdty\u002Fpurple)\n1. METASQL: A Generate-then-Rank Framework for Natural Language to SQL Translation. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2024-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.17144) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FKaimary\u002FMetaSQL)\n1. Archer: A Human-Labeled Text-to-SQL Dataset with Arithmetic, Commonsense and Hypothetical Reasoning.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2024-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2024.eacl-long.6\u002F) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fsig4kg.github.io\u002Farcher-bench\u002F)\n1. Synthesizing Text-to-SQL Data from Weak and Strong LLMs.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2024-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2408.03256) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FYangjiaxi\u002FSense)\n1. Understanding the Effects of Noise in Text-to-SQL: An Examination of the BIRD-Bench Benchmark.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2024-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2402.12243) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fniklaswretblad\u002Fthe-effects-of-noise-in-text-to-SQL)\n1. I Need Help! Evaluating LLM’s Ability to Ask for Users’ Support: A Case Study on Text-to-SQL Generation.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2024-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2407.14767) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fappier-research\u002Fi-need-help)\n1. PTD-SQL: Partitioning and Targeted Drilling with LLMs in Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2024-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2409.14082) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Flrlbbzl\u002FPTD-SQL)\n1. Improving Retrieval-augmented Text-to-SQL with AST-based Ranking and Schema Pruning.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2024-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2407.03227)\n1. Data-Centric Text-to-SQL with Large Language Models. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNeurIPS(workshop)'2024-yellow\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fpdf?id=gDKIjZcg93)\n1. Research and Practice on Database Interaction Based on Natural Language Processing\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAIAC'2024-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.17894)\n1. XiYan-SQL: A Multi-Generator Ensemble Framework for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2411.08599)\n1. Structure Guided Large Language Model for SQL Generation. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2402.13284) \n1. A Plug-and-Play Natural Language Rewriter for Natural Language to SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2412.17068) \n1. RSL-SQL: Robust Schema Linking in Text-to-SQL Generation.   \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.15879) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fglee4810\u002FTrustSQL)\n1. In-Context Reinforcement Learning based Retrieval-Augmented Generation for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fassets.amazon.science\u002F09\u002Ff4\u002F493c574346f895bbb0303282a501\u002Fin-context-reinforcement-learning-based-retrieval-augmented-generation-for-text-to-sql.pdf) \n1. TrustSQL: Benchmarking Text-to-SQL Reliability with Penalty-Based Scoring.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2411.00073) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FLaqcce-cao\u002FRSL-SQL)\n1. LAIA-SQL: Enhancing Natural Language to SQL Generation in Multi-Table QA via Task Decomposition and Keyword Extraction\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fpdf?id=WYdpjwKQma)\n1. Research on Large Model Text-to-SQL Optimization Method for Intelligent Interaction in the Field of Construction Safety.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F10810146)\n1. SQLh-GEN: Bridging the Dialect Gap for Text-to-SQL Via Synthetic Data And Model Merging.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2408.12733v2)\n1. Grounding Natural Language to SQL Translation with Data-Based Self-Explanations.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2411.02948) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FKaimary\u002FCycleSQL)\n1. Towards Optimizing SQL Generation via LLM Routing.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.04319)\n1. E-SQL: Direct Schema Linking via Question Enrichment in Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.16751) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FHasanAlpCaferoglu\u002FE-SQL)\n1. DB-GPT: Empowering Database Interactions with Private Large Language Models.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.17449) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Feosphoros-ai\u002FDB-GPT)\n1. The Death of Schema Linking? Text-to-SQL in the Age of Well-Reasoned Language Models.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2408.07702)  \n1. CHESS: Contextual Harnessing for Efficient SQL Synthesis.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.16755) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FShayanTalaei\u002FCHESS)\n1. PET-SQL: A Prompt-Enhanced Two-Round Refinement of Text-to-SQL with Cross-consistency.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.09732) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fruc-datalab\u002FZeroNL2SQL)\n1. CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.02712) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FX-LANCE\u002Ftext2sql-multiturn-GPT)\n1. AMBROSIA: A Benchmark for Parsing Ambiguous Questions into Database Queries.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.19073) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fambrosia-benchmark.github.io\u002F)\n1. Text-to-SQL Calibration: No Need to Ask—Just Rescale Model Probabilities.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2411.16742) \n1. Few-shot Text-to-SQL Translation using Structure and Content Prompt Learning.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2023-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3589292) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fruc-datalab\u002FSC-prompt)\n1. CatSQL: Towards Real World Natural Language to SQL Applications.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2023-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol16\u002Fp1534-fu.pdf) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fasfuhan\u002FCatSQL)\n1. DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correction. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNeurIPS'2023-yellow\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.11015) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FMohammadrezaPourreza\u002FFew-shot-NL2SQL-with-prompting\u002Ftree\u002Fmain)\n1. Data Ambiguity Strikes Back: How Documentation Improves GPT's Text-to-SQL. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNeurIPS(workshop)'2023-yellow\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fpdf?id=FflKTuIRTD) \n1. ACT-SQL: In-Context Learning for Text-to-SQL with Automatically-Generated Chain-of-Thought.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2023-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.17342) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FX-LANCE\u002Ftext2sql-GPT)\n1. Selective Demonstrations for Cross-domain Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2023-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.06302) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fshuaichenchang\u002FODIS-Text-to-SQL)\n1. RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQL. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2023-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2302.05965) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRUCKBReasoning\u002FRESDSQL)\n1. Graphix-T5: Mixing Pre-trained Transformers with Graph-Aware Layers for Text-to-SQL Parsing. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2023-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.07507) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FAlibabaResearch\u002FDAMO-ConvAI\u002Ftree\u002Fmain\u002Fgraphix)\n1. Improving Generalization in Language Model-based Text-to-SQL Semantic Parsing: Two Simple Semantic Boundary-based Techniques.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2023-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fvirtual2023.aclweb.org\u002Fpaper_P4350.html) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FDakingrai\u002Food-generalization-semantic-boundary-techniques)\n1. G\u003Csup>3\u003C\u002Fsup>R: A Graph-Guided Generate-and-Rerank Framework for Complex and Cross-domain Text-to-SQL Generation.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(findings)'2023-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2023.findings-acl.23\u002F) \n1. Importance of Synthesizing High-quality Data for Text-to-SQL Parsing.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(findings)'2023-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2023.findings-acl.86.pdf) \n1. Know What I don’t Know: Handling Ambiguous and Unknown Questions for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(findings)'2023-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2023.findings-acl.352\u002F) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fwbbeyourself\u002FDTE)\n1. C3: Zero-shot Text-to-SQL with ChatGPT \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2023-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.07306) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fbigbigwatermalon\u002FC3SQL)\n1. SQLformer: Deep Auto-Regressive Query Graph Generation for Text-to-SQL Translation.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2023-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.18376) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FAdrianBZG\u002FSQLformer)\n\n\n## 📊 Text-to-SQL Benchmark\nWe create a timeline of the benchmark's development and mark relevant milestones. You can get more details from this chapter: [📊 Benchmark](chapter\u002FBenchmark.md)\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002FDataset_timeline.svg\"\u002F>\n\u003C\u002Fp>\n\n## 🎯 Where Are We Going?\n\n* 🎯Solve Open Text-to-SQL Problem\n* 🎯Develop Cost-effective Text-to-SQL Methods\n* 🎯Make Text-to-SQL Solutions Trustworthy\n* 🎯Text-to-SQL with Ambiguous and Unspecified NL Queries\n* 🎯Adaptive Training Data Synthesis\n\n## 📖 Catalog for Our Survey\nYou can get more information from our subsection. We introduce representative papers on related concepts:\n* [Pre-Processing](chapter\u002FPre_Processing.md)\n* [Text-to-SQL Translation Methods](chapter\u002FTranslation_method.md)\n* [Post-Processing](chapter\u002FPost_Processing.md)\n* [Benchmark](chapter\u002FBenchmark.md)\n* [Evaluation](chapter\u002FEvaluation.md)\n* [Error Analysis](chapter\u002FError_Analysis.md)\n\n## 💾 Practical Guide for Novice\n\n### 📊 How to get data:\n* We collect Text-to-SQL benchmark features and download links for you. You can get more details from this chapter: [Benchmark](chapter\u002FBenchmark.md)\n* The analysis code for benchmarks is available in the `src\u002Fdataset_analysis` directory. Benchmark analysis reports can be found in the `report\u002F` directory.\n\n### 🛠️ How to build an LLM-based Text-to-SQL model:\n\n* Litgpt [Repository Link](https:\u002F\u002Fgithub.com\u002FLightning-AI\u002Flitgpt)\n\n    This repository offers access to over 20 high-performance large language models (LLMs) with comprehensive guides for pretraining, fine-tuning, and deploying at scale. It is designed to be beginner-friendly with from-scratch implementations and no complex abstractions.\n\n* LLaMA-Factory [Repository Link](https:\u002F\u002Fgithub.com\u002Fhiyouga\u002FLLaMA-Factory)\n    Unified Efficient Fine-Tuning of 100+ LLMs. Integrating various models with scalable training resources, advanced algorithms, practical tricks, and comprehensive experiment monitoring tools, this setup enables efficient and faster inference through optimized APIs and UIs.\n\n* Fine-tuning and In-Context learning for BIRD-SQL benchmark [Repository Link](https:\u002F\u002Fgithub.com\u002FAlibabaResearch\u002FDAMO-ConvAI\u002Ftree\u002Fmain\u002Fbird#fine-tuning-ft)\n    \n    A tutorial for both Fine-tuning and In-Context Learning is provided by the BIRD-SQL benchmark. \n\n### 🔎How to evaluate your model:\n\nWe collect NL2SQL evaluation metrics for you. You can get more details from this chapter: [Evaluation](chapter\u002FEvaluation.md)\n\n* NLSQL360 [Repository Link](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL360) \n\n     NL2SQL360 is a testbed for fine-grained evaluation of NL2SQL solutions. Our testbed integrates existing NL2SQL benchmarks, a repository of NL2SQL models, and various evaluation metrics, which aims to provide an intuitive and user-friendly platform to enable both standard and customized performance evaluations. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEX-red\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEM-green\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVES-blue\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FQVT-orange\">\n\n* Test-suite-sql-eval [Repository Link](https:\u002F\u002Fgithub.com\u002Ftaoyds\u002Ftest-suite-sql-eval)\n\n    This repo contains a test suite evaluation metric for 11 text-to-SQL tasks. It is now the official metric of [Spider](https:\u002F\u002Fyale-lily.github.io\u002Fspider), [SParC](https:\u002F\u002Fyale-lily.github.io\u002Fsparc), and [CoSQL](https:\u002F\u002Fyale-lily.github.io\u002Fcosql), and is also now available for Academic, ATIS, Advising, Geography, IMDB, Restaurants, Scholar, and Yelp (building on the amazing work by [Catherine and Jonathan](https:\u002F\u002Fgithub.com\u002Fjkkummerfeld\u002Ftext2sql-data)).  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEX-red\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEM-green\">\n\n* BIRD-SQL-Official [Repository Link](https:\u002F\u002Fgithub.com\u002FAlibabaResearch\u002FDAMO-ConvAI\u002Ftree\u002Fmain\u002Fbird#evaluation)\n\n    It is now the official tool of [BIRD-SQL](https:\u002F\u002Fbird-bench.github.io\u002F). It is the first tool to propose VES and give an official test suite. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEX-red\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVES-blue\">\n\n\n### 🗺️ Roadmap and Decision Flow\n\nYou can get some inspiration from the Roadmap and Decision Flow.\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002FNL2SQL_Guidance.svg\"\u002F>\n\u003C\u002Fp>\n\n## 📱 Text-to-SQL Related Applications:\n\n* AI for Database: Agentic AI product for databases — connect any database (PostgreSQL, MySQL, MongoDB, etc.) and talk to it in plain English. Features self-refreshing intelligent dashboards, natural language queries, and automated action workflows that trigger on database changes. [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb Link-98f\">](https:\u002F\u002Faifordatabase.com)\n* Chat2DB: AI-driven database tool and SQL client, The hottest GUI client, supporting MySQL, Oracle, PostgreSQL, DB2, SQL Server, DB2, SQLite, H2, ClickHouse, and more. [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRepositor Link-grey\">](https:\u002F\u002Fgithub.com\u002FcodePhiliaX\u002FChat2DB) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb Link-98f\">](https:\u002F\u002Fchat2db-ai.com\u002Fzh-CN)\n* DB-GPT: AI Native Data App Development framework with AWEL(Agentic Workflow Expression Language) and Agents. [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRepositor Link-grey\">](https:\u002F\u002Fgithub.com\u002Feosphoros-ai\u002FDB-GPT) \n* Postgres.new: In-browser Postgres sandbox with AI assistance.  [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRepositor Link-grey\">](https:\u002F\u002Fgithub.com\u002Fsupabase-community\u002Fpostgres-new\u002Ftree\u002Fmain) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb Link-98f\">](https:\u002F\u002Fpostgres.new\u002F)\n* QueryGPT – Natural Language to SQL Using Generative AI. [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb Link-98f\">](https:\u002F\u002Fwww.uber.com\u002Fen-JP\u002Fblog\u002Fquery-gpt\u002F)\n\n## 📮Connect with Us\nPlease feel free to contact us if we missed any interesting work.\n\n📧 xliu371[at]connect.hkust-gz.edu.cn\n\n","\u003Ch1 align=\"center\">文本转SQL手册\u003C\u002Fh1>\n\n \u003Ch3 align=\"center\">NL2SQL手册\u003C\u002Fh3>\n \n这是**[TKDE'25] 大语言模型时代的文本转SQL综述：我们目前处于什么阶段，未来又将走向何方？**和**[VLDB'24] 自然语言转SQL的曙光：我们准备好了吗？**的官方仓库。\n通过该仓库，您可以探索文本转SQL研究（即NL2SQL）领域的[最新进展](#-text-to-sql-survey--tutorial)。我们提供了全面的综述、深入的研究论文以及基准评估。\n\n**\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2025-green\"> 大语言模型时代的文本转SQL综述：我们目前处于什么阶段，未来又将走向何方？**\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.05109)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSlides-orange\">](.\u002Fslides\u002FNL2SQL_handbook.pdf)\n\n**\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> 自然语言转SQL：现状与开放问题**\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdbgroup.cs.tsinghua.edu.cn\u002Fligl\u002Fpapers\u002FVLDB25-NL2SQL.pdf)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSlides-orange\">](.\u002Fslides\u002FNL2SQL-VLDB2025.pdf)\n\n**\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> 自然语言转SQL的曙光：我们准备好了吗？**\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol17\u002Fp3318-luo.pdf) \n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSlides-orange\">](.\u002Fslides\u002FNL2SQL360-VLDB2024.pdf)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-purple\">](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL360)\n\n📧如果您认为我们遗漏了任何有趣的工作，请[联系我们](#connect-with-us)。\n\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002Friver.svg\"\u002F>\n\u003C\u002Fp>\n\n```bibtex\n@article{liu2025survey,\n  title={大语言模型时代的文本转SQL综述：我们目前处于什么阶段，未来又将走向何方？},\n  author={刘欣宇、沈书宇、李博彦、马培贤、蒋润志、张宇鑫、范炬、李国梁、唐楠、罗雨雨},\n  journal={IEEE知识与数据工程汇刊},\n  year={2025},\n  publisher={IEEE}\n}\n```\n\n## 🧭 文本转SQL简介 \n将用户的自然语言查询（NL）转换为SQL查询，可以显著降低访问关系型数据库的门槛，并支持多种商业应用。随着语言模型（LMs）的出现，文本转SQL的性能得到了极大提升。在此背景下，评估当前的发展状况、确定从业者在特定场景下应采用的文本转SQL解决方案，以及明确研究人员下一步的研究方向，显得尤为重要。\n\n\u003Cp align=\"center\">\n\u003Cimg width=\"600\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FHKUSTDial_NL2SQL_Handbook_readme_8b45a8d62150.jpg\"\u002F>\n\u003C\u002Fp>\n\n## 📈 文本转SQL生命周期\n\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002Fnl2sql_lifecycle.svg\"\u002F>\n\u003C\u002Fp>\n\n+ 模型：文本转SQL翻译技术不仅要应对自然语言的歧义和信息不足问题，还要正确地将自然语言与数据库模式和实例进行映射；\n+ 数据：从训练数据的收集、因训练数据稀缺而进行的数据合成，到文本转SQL基准测试；\n+ 评估：使用不同指标和粒度从多个角度评估文本转SQL方法；\n+ 错误分析：分析文本转SQL中的错误以找出根本原因，并指导文本转SQL模型不断改进。\n\n## 🤔 我们目前处于什么阶段？\n我们将文本转SQL面临的挑战分为五个层次，每一层都针对特定的难题。前三个层次涵盖了已经或正在解决的挑战，反映了文本转SQL的逐步发展。第四层次代表我们在大语言模型阶段希望攻克的挑战，而第五层次则勾勒出我们在未来五年内对文本转SQL系统的愿景。\n我们从语言模型的角度描述了文本转SQL解决方案的演进过程，并将其划分为四个阶段。\n对于文本转SQL的每个阶段，我们都分析了目标用户的变化以及挑战被解决的程度。\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002F从语言模型角度看NL2SQL解决方案的演进.svg\"\u002F>\n\u003C\u002Fp>\n\n\n## 🧩 基于模块的文本转SQL方法 \n我们总结了利用语言模型的文本转SQL解决方案中的关键模块。\n+ **预处理**是在文本转SQL解析过程中对模型输入的增强。您可以在本章中了解更多详情：[预处理](chapter\u002FPre_Processing.md)\n+ **文本转SQL翻译方法**是文本转SQL解决方案的核心，负责将输入的自然语言查询转换为SQL查询。您可以在本章中了解更多详情：[文本转SQL翻译方法](chapter\u002FTranslation_method.md)\n+ **后处理**是优化生成的SQL查询的关键步骤，以确保其更准确地满足用户期望。您可以在本章中了解更多详情：[后处理](chapter\u002FPost_Processing.md)\n\u003Cp align=\"center\">\n\u003Cimg width=\"600\" src=\".\u002Fassets\u002F大语言模型时代NL2SQL方法概述.svg\"\u002F>\n\u003C\u002Fp>\n\n## 📚 Text-to-SQL 综述与教程\n\n1. 大型语言模型时代的 Text-to-SQL 综述：我们身处何处，又将走向何方？\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.05109) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL_Handbook)\n1. 自然语言到 SQL：现状与开放问题。 \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdbgroup.cs.tsinghua.edu.cn\u002Fligl\u002Fpapers\u002FVLDB25-NL2SQL.pdf)\n1. 下一代数据库接口：基于大语言模型的 Text-to-SQL 综述。\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.08426)\n1. 利用大型语言模型处理 Text-to-SQL 任务的综述。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCSUR'2024-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2407.15186)\n1. 大型语言模型增强的 Text-to-SQL 生成：综述。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.06011)\n1. 从自然语言到 SQL：基于大语言模型的 Text-to-SQL 系统综述。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2410.01066)\n1. 面向表格数据查询和可视化的自然语言接口：综述。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2024-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.17894)\n1. 基于深度学习的数据库自然语言接口。\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2023-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.14778\u002F3611540.3611575)\n1. 关于 Text-to-SQL 的深度学习方法综述。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDBJ'2023-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1007\u002Fs00778-022-00776-8)\n1. Text-to-SQL 解析综述：概念、方法与未来方向。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTKDE'2022-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2208.13629)\n1. Text-to-SQL 的最新进展：现有成果与未来展望综述。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLING'2022-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2022.coling-1.190\u002F)\n1. 深入探讨 Text-to-SQL 系统中的深度学习方法。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2021-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1145\u002F3448016.3457543)\n1. 数据自然语言接口的现状与开放挑战。\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2020-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1145\u002F3318464.3383128)\n1. 自然语言到 SQL：我们目前处于什么阶段？ \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2020-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol13\u002Fp1737-kim.pdf)\n\n## 📰 Text-to-SQL Paper List\n1. DeepEye-SQL: A Software-Engineering-Inspired Text-to-SQL Framework. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2026-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.17586)\n1. Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Search\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICML'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.17248) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Falpha-sql-hkust.github.io\u002F)\n1. NL2SQL-BUGs: A Benchmark for Detecting Semantic Errors in NL2SQL Translation.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGKDD'2025-B6FFBB\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.11984) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fnl2sql-bugs.github.io\u002F)\n1. EllieSQL: Cost-Efficient Text-to-SQL with Complexity-Aware Routing. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLM'2025-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.22402) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Felliesql.github.io\u002F)\n1. The Dawn of Natural Language to SQL: Are We Fully Ready?\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.01265) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL360)\n2. DIVER: A Robust Text-to-SQL System with Dynamic Interactive Value Linking and Evidence Reasoning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2026-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2602.12064)\n3. LogicCat: A Chain-of-Thought Text-to-SQL Benchmark for Complex Reasoning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2026-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fojs.aaai.org\u002Findex.php\u002FAAAI\u002Farticle\u002Fview\u002F40243)\n4. Schema on the Inside: A Two-Phase Fine-Tuning Method for High-Efficiency Text-to-SQL at Scale. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2026-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fojs.aaai.org\u002Findex.php\u002FAAAI\u002Farticle\u002Fview\u002F41446)\n5. Hexgen-Flow: Optimizing LLM Inference Request Scheduling for Agentic Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.05286) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRelaxed-System-Lab\u002FHexgen-Flow)\n6. SQLMorph: Query Mutation and Fine-Grained Metrics for Text-to-SQL Evaluation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Famine.io\u002Fpapers\u002F2026-icde-sqlmorph.pdf)  [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fdais-polymtl\u002Fsqlmorph)\n7. LEAF-SQL: Level-wise Exploration with Adaptive Fine-graining for Text-to-SQL Skeleton Prediction. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\">  [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FAtlamtiz\u002FLEAF-SQL)\n8. An Efficient and Effective Evaluator for Text2SQL Models on Unseen and Unlabeled Data. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\">\n9. Boosting Small Language Models for Text-to-SQL with Fine-Grained Execution Feedback and Cost-Efficient Rewards. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\">\n10. MM2SQL: A Benchmark and Method for Visually-Grounded SQL Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> \n11. Text2SQL-Flow: A Robust SQL-Aware Data Augmentation Framework for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.10192)\n12. Knapsack Optimization-based Schema Linking for LLM-based Text-to-SQL Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.12911)\n13. CYANSQL: Unlock the Power of NL2SQL via Clustering-based Test-Time Scaling. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2026-green\">\n14. Beyond Static Pipelines: Learning Dynamic Workflows for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.15077) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FSatissss\u002FSquRL)\n15. ReViSQL: Achieving Human-Level Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.20004) \n16. Think2SQL: Blueprinting Reward Density and Advantage Scaling for Effective Text-to-SQL Reasoning.  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">]([https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.15077](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2602.15564)) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fanonymous.4open.science\u002Fr\u002FThink2SQL-3B7F\u002FREADME.md)\n1. AgentSM: Semantic Memory for Agentic Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2601.15709)\n1. LLM-Based SQL Generation: Prompting, Self-Refinement, and Adaptive Weighted Majority Voting. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2601.17942)\n1. Pervasive Annotation Errors Break Text-to-SQL Benchmarks and Leaderboards.  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.08778) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fuiuc-kang-lab\u002Ftext_to_sql_benchmarks)\n1. OptiSQL: Executable SQL Generation from Optical Tokens. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2601.13695) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FjohnnyZeppelin\u002FOptiSQL) \n1. Memo-SQL: Structured Decomposition and Experience-Driven Self-Correction for Training-Free NL2SQL.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2026-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.arxiv.org\u002Fpdf\u002F2601.10011)\n1. Structure-Guided Large Language Models for Text-to-SQL Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICML'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Ficml.cc\u002Fvirtual\u002F2025\u002Fposter\u002F44477)\n1. Sphinteract: Resolving Ambiguities in NL2SQL Through User Interaction.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol18\u002Fp1145-zhao.pdf) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FZhaoFuheng\u002FSphinteract)\n1. OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scale. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">]([https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol18\u002Fp1145-zhao.pdf](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.02240)) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRUCKBReasoning\u002FOmniSQL)\n1. EVOSCHEMA: TOWARDS TEXT-TO-SQL ROBUSTNESS AGAINST SCHEMA EVOLUTION. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fpdf?id=NfUHBaZdLw) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fzhangtianshu\u002FEvoSchema)\n1. Is Long Context All You Need? Leveraging LLM's Extended Context for NL2SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.12372) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fyeounoh\u002Flc_nl2sql)\n1. The Power of Constraints in Natural Language to SQL Translation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2025-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol18\u002Fp2097-ren.pdf)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fhttdty\u002FREDSQL_VLDB)\n1. OpenSearch-SQL: Enhancing Text-to-SQL with Dynamic Few-shot and Consistency Alignment. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2025-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.14913) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FOpenSearch-AI\u002FOpenSearch-SQL)\n1. Reliable Text-to-SQL with Adaptive Abstention.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2025-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.10858) \n1. SNAILS: Schema Naming Assessments for Improved LLM-Based SQL Inference.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2025-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1145\u002F3709727)\n1. Automated Validating and Fixing of Text-to-SQL Translation with Execution Consistency. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2025-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fipads.se.sjtu.edu.cn\u002Fzh\u002Fpublications\u002FSQLDriller.pdf)\n1. Grounding Natural Language to SQL Translation with Data-Based Self-Explanations.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.02948) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FKaimary\u002FCycleSQL)\n1. AID-SQL: Adaptive In-Context Learning of Text-to-SQL with Difficulty-Aware Instruction and Retrieval-Augmented Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.computer.org\u002Fcsdl\u002Fproceedings-article\u002Ficde\u002F2025\u002F360300d945\u002F26FZCc99mg0) \n1. CLEAR: A Parser-Independent Disambiguation Framework for NL2SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2025-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.computer.org\u002Fcsdl\u002Fproceedings-article\u002Ficde\u002F2025\u002F360300d302\u002F26FZBD2hBJe) \n1. CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICLR'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2410.01943v1) \n1. Spider 2.0: Evaluating Language Models on Real-World Enterprise Text-to-SQL Workflows.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICLR'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.07763) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fxlang-ai\u002FSpider2)\n1. ROUTE: Robust Multitask Tuning and Collaboration for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICLR'2025-brightgreen\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2412.10138)\n1. SHARE: An SLM-based Hierarchical Action CorREction Assistant for Text-to-SQL.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.00391)\n1. DCG-SQL: Enhancing In-Context Learning for Text-to-SQL with Deep Contextual Schema Link Graph. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.19956)\n1. Uncovering the Impact of Chain-of-Thought Reasoning for Direct Preference Optimization: Lessons from Text-to-SQL.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.11656)\n1. STaR-SQL: Self-Taught Reasoner for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13550)\n1. SQLGenie: A Practical LLM based System for Reliable and Efficient SQL Generation \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2025(industry)-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">]([https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13550](https:\u002F\u002Faclanthology.org\u002F2025.acl-industry.71\u002F))\n1. SQL-R1: Training Natural Language to SQL Reasoning Model By Reinforcement Learning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNeurIPS'2025-yellow\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.08600) \n1. Confidence Estimation for Error Detection in Text-to-SQL Systems. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2025-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.09527)\n1. SQLord: A Robust Enterprise Text-to-SQL Solution via Reverse Data Generation and Workflow Decomposition. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWWW'2025-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fpdf\u002F10.1145\u002F3701716.3715541)\n1. DBCopilot: Scaling Natural Language Querying to Massive Databases.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEDBT\u002FICDT'2025-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.03463) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Ftshu-w\u002FDBCopilot)\n1. Utilising Large Language Models for Adversarial Attacks in Text-to-SQL: A Perpetrator and Victim Approach.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBTW'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.20657) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FXGenerationLab\u002FXiYan-DBDescGen)\n1. You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNAACL'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.12172) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fsig4kg.github.io\u002Farcher-bench\u002F)\n1. Boosting Text-to-SQL through Multi-grained Error Identification.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLING'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2025.coling-main.289.pdf)\n1. Gen-SQL: Efficient Text-to-SQL By Bridging Natural Language Question And Database Schema With Pseudo-Schema.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLING'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2025.coling-main.256\u002F)\n1. MAC-SQL: A Multi-Agent Collaborative Framework for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLING'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.11242) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fwbbeyourself\u002FMAC-SQL)\n1. PARSQL: Enhancing Text-to-SQL through SQL Parsing and Reasoning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2024.findings-acl.120\u002F)\n1. UCS-SQL: Uniting Content and Structure for Enhanced Semantic Bridging In Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fforum?id=xnTouV7wyr)\n1. SQLForge: Synthesizing Reliable and Diverse Data to Enhance Text-to-SQL Reasoning in LLMs. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.13725)\n1. Optimizing Reasoning for Text-to-SQL with Execution Feedback. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.19988)\n1. Knowledge Base Construction for Knowledge-Augmented Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Findings)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2505.22096)\n1. SQLong: Enhanced NL2SQL for Longer Contexts with LLMs.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(Workshop)'2025-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.16747)\n1. Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCOLM'2025-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.23157)\n1. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBIRD Top1-red\">Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scaling.  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.24403).\n1. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBIRD Top2-blue\">Automatic Metadata Extraction for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.19988)\n1. DeepEye-SQL: A Software-Engineering-Inspired Text-to-SQL Framework. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.17586)\n1. Rethinking Text-to-SOL: Dynamic Multi-turn SOIInteraction for Real-world Database Exploration. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2510.26495) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FAurora-slz\u002FRealWorld-SQL-Bench)\n1. MARS-SQL: A MULTI-AGENT REINFORCEMENT LEARNING FRAMEWORK FOR TEXT-TO-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2511.01008) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FYangHaolin0526\u002FMARS-SQL)\n1. RUBIKSQL: Lifelong Learning Agentic Knowledge Base as an Industrial NL2SQL System. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2508.17590) \n1. CSC-SQL: Corrective Self-Consistency in Text-to-SQL via Reinforcement Learning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.13271) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FCycloneBoy\u002Fcsc_sql\u002F)\n1. Cheaper, Better, Faster, Stronger: Robust Text-to-SQL without Chain-of-Thought or Fine-Tuning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2505.14174) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fgenaasia\u002FN-rep)\n1. SLM-SQL: An Exploration of Small Language Models for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.22478) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FCycloneBoy\u002Fslm_sql)\n1. Reward-SQL: Boosting Text-to-SQL via Stepwise Reasoning and Process-Supervised Rewards. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2505.04671) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fruc-datalab\u002FRewardSQL)\n1. Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.20315) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fsnowflakedb\u002FArcticTraining)\n1. Corrective Self-Consistency in Text-to-SQL via Reinforcement Learning. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.13271) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FCycloneBoy\u002Fcsc_sql)\n1. SQLForge: Synthesizing Reliable and Diverse Data to Enhance Text-to-SQL Reasoning in LLMs. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2505.13725)\n1. Think2SQL: Reinforce LLM Reasoning Capabilities for Text2SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.15077)\n1. Distill-C: Enhanced NL2SQL via Distilled Customization with LLMs. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.00048)\n1. OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scale. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.02240) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRUCKBReasoning\u002FOmniSQL)\n1. SQL-Factory: A Multi-Agent Framework for High-Quality and Large-Scale SQL Generation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.14837)\n1. Text2SQL is Not Enough: Unifying AI and Databases with TAG. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2408.14717) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FTAG-Research\u002FTAG-Bench) \n1. Automatic database description generation for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.20657)\n[\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FXGenerationLab\u002FXiYan-DBDescGen)\n1. MCTS-SQL: An Effective Framework for Text-to-SQL with Monte Carlo Tree Search.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.16607)\n1. SQL-o1: A Self-Reward Heuristic Dynamic Search Method for Text-to-SQL. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2502.11741)\n1. FEATHER-SQL: A Lightweight NL2SQL Framework with Dual-Model Collaboration Paradigm for Small Language Models.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.17811)\n1. FI-NL2PY2SQL: Financial Industry NL2SQL Innovation Model Based on Python and Large Language Model.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F17\u002F1\u002F12)\n1. FGCSQL: A Three-Stage Pipeline for Large Language Model-Driven Chinese Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.mdpi.com\u002F2079-9292\u002F14\u002F6\u002F1214)\n1. Transforming Medical Data Access: The Role and Challenges of Recent Language Models in SQL Query Automation. \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2025-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.mdpi.com\u002F1999-4893\u002F18\u002F3\u002F124)\n1. The Dawn of Natural Language to SQL: Are We Fully Ready?\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.01265) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL360)\n1. Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.15363) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FBeachWang\u002FDAIL-SQL) \n1. Interleaving Pre-Trained Language Models and Large Language Models for Zero-Shot NL2SQL Generation. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.08891) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fruc-datalab\u002FZeroNL2SQL)\n1. Generating Succinct Descriptions of Database Schemata for Cost-Efficient Prompting of Large Language Models. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.14778\u002F3681954.3682017) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fitrummer\u002Fschemacompression)\n1. ScienceBenchmark: A Complex Real-World Benchmark for Evaluating Natural Language to SQL Systems.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2024-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.04743) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fsciencebenchmark.cloudlab.zhaw.ch\u002F)\n1. CodeS: Towards Building Open-source Language Models for Text-to-SQL. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2024-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.16347) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRUCKBReasoning\u002Fcodes)\n1. FinSQL: Model-Agnostic LLMs-based Text-to-SQL Framework for Financial Analysis. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSIGMOD'2024-red\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.10506) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fbigbigwatermalon\u002FFinSQL)\n1. PURPLE: Making a Large Language Model a Better SQL Writer. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2024-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.20014) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fhttdty\u002Fpurple)\n1. METASQL: A Generate-then-Rank Framework for Natural Language to SQL Translation. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FICDE'2024-green\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.17144) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FKaimary\u002FMetaSQL)\n1. Archer: A Human-Labeled Text-to-SQL Dataset with Arithmetic, Commonsense and Hypothetical Reasoning.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2024-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2024.eacl-long.6\u002F) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fsig4kg.github.io\u002Farcher-bench\u002F)\n1. Synthesizing Text-to-SQL Data from Weak and Strong LLMs.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2024-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2408.03256) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FYangjiaxi\u002FSense)\n1. Understanding the Effects of Noise in Text-to-SQL: An Examination of the BIRD-Bench Benchmark.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2024-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2402.12243) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fniklaswretblad\u002Fthe-effects-of-noise-in-text-to-SQL)\n1. I Need Help! Evaluating LLM’s Ability to Ask for Users’ Support: A Case Study on Text-to-SQL Generation.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2024-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2407.14767) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fappier-research\u002Fi-need-help)\n1. PTD-SQL: Partitioning and Targeted Drilling with LLMs in Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2024-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2409.14082) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Flrlbbzl\u002FPTD-SQL)\n1. Improving Retrieval-augmented Text-to-SQL with AST-based Ranking and Schema Pruning.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2024-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2407.03227)\n1. Data-Centric Text-to-SQL with Large Language Models. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNeurIPS(workshop)'2024-yellow\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fpdf?id=gDKIjZcg93)\n1. Research and Practice on Database Interaction Based on Natural Language Processing\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAIAC'2024-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.17894)\n1. XiYan-SQL: A Multi-Generator Ensemble Framework for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2411.08599)\n1. Structure Guided Large Language Model for SQL Generation. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2402.13284) \n1. A Plug-and-Play Natural Language Rewriter for Natural Language to SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2412.17068) \n1. RSL-SQL: Robust Schema Linking in Text-to-SQL Generation.   \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.15879) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fglee4810\u002FTrustSQL)\n1. In-Context Reinforcement Learning based Retrieval-Augmented Generation for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fassets.amazon.science\u002F09\u002Ff4\u002F493c574346f895bbb0303282a501\u002Fin-context-reinforcement-learning-based-retrieval-augmented-generation-for-text-to-sql.pdf) \n1. TrustSQL: Benchmarking Text-to-SQL Reliability with Penalty-Based Scoring.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2411.00073) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FLaqcce-cao\u002FRSL-SQL)\n1. LAIA-SQL: Enhancing Natural Language to SQL Generation in Multi-Table QA via Task Decomposition and Keyword Extraction\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fpdf?id=WYdpjwKQma)\n1. Research on Large Model Text-to-SQL Optimization Method for Intelligent Interaction in the Field of Construction Safety.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F10810146)\n1. SQLh-GEN: Bridging the Dialect Gap for Text-to-SQL Via Synthetic Data And Model Merging.\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2408.12733v2)\n1. Grounding Natural Language to SQL Translation with Data-Based Self-Explanations.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2411.02948) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FKaimary\u002FCycleSQL)\n1. Towards Optimizing SQL Generation via LLM Routing.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.04319)\n1. E-SQL: Direct Schema Linking via Question Enrichment in Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.16751) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FHasanAlpCaferoglu\u002FE-SQL)\n1. DB-GPT: Empowering Database Interactions with Private Large Language Models.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.17449) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Feosphoros-ai\u002FDB-GPT)\n1. The Death of Schema Linking? Text-to-SQL in the Age of Well-Reasoned Language Models.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2408.07702)  \n1. CHESS: Contextual Harnessing for Efficient SQL Synthesis.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.16755) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FShayanTalaei\u002FCHESS)\n1. PET-SQL: A Prompt-Enhanced Two-Round Refinement of Text-to-SQL with Cross-consistency.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.09732) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fruc-datalab\u002FZeroNL2SQL)\n1. CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.02712) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FX-LANCE\u002Ftext2sql-multiturn-GPT)\n1. AMBROSIA: A Benchmark for Parsing Ambiguous Questions into Database Queries.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.19073) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fambrosia-benchmark.github.io\u002F)\n1. Text-to-SQL Calibration: No Need to Ask—Just Rescale Model Probabilities.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2024-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2411.16742) \n1. Few-shot Text-to-SQL Translation using Structure and Content Prompt Learning.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2023-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3589292) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fruc-datalab\u002FSC-prompt)\n1. CatSQL: Towards Real World Natural Language to SQL Applications.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVLDB'2023-blue\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fwww.vldb.org\u002Fpvldb\u002Fvol16\u002Fp1534-fu.pdf) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fasfuhan\u002FCatSQL)\n1. DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correction. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNeurIPS'2023-yellow\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.11015) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FMohammadrezaPourreza\u002FFew-shot-NL2SQL-with-prompting\u002Ftree\u002Fmain)\n1. Data Ambiguity Strikes Back: How Documentation Improves GPT's Text-to-SQL. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNeurIPS(workshop)'2023-yellow\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fopenreview.net\u002Fpdf?id=FflKTuIRTD) \n1. ACT-SQL: In-Context Learning for Text-to-SQL with Automatically-Generated Chain-of-Thought.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2023-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.17342) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FX-LANCE\u002Ftext2sql-GPT)\n1. Selective Demonstrations for Cross-domain Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEMNLP'2023-orange\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.06302) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fshuaichenchang\u002FODIS-Text-to-SQL)\n1. RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQL. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2023-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2302.05965) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FRUCKBReasoning\u002FRESDSQL)\n1. Graphix-T5: Mixing Pre-trained Transformers with Graph-Aware Layers for Text-to-SQL Parsing. \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAAAI'2023-cyan\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.07507) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FAlibabaResearch\u002FDAMO-ConvAI\u002Ftree\u002Fmain\u002Fgraphix)\n1. Improving Generalization in Language Model-based Text-to-SQL Semantic Parsing: Two Simple Semantic Boundary-based Techniques.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL'2023-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Fvirtual2023.aclweb.org\u002Fpaper_P4350.html) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FDakingrai\u002Food-generalization-semantic-boundary-techniques)\n1. G\u003Csup>3\u003C\u002Fsup>R: A Graph-Guided Generate-and-Rerank Framework for Complex and Cross-domain Text-to-SQL Generation.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(findings)'2023-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2023.findings-acl.23\u002F) \n1. Importance of Synthesizing High-quality Data for Text-to-SQL Parsing.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(findings)'2023-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2023.findings-acl.86.pdf) \n1. Know What I don’t Know: Handling Ambiguous and Unknown Questions for Text-to-SQL.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FACL(findings)'2023-9cf\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Faclanthology.org\u002F2023.findings-acl.352\u002F) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fwbbeyourself\u002FDTE)\n1. C3: Zero-shot Text-to-SQL with ChatGPT \n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2023-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.07306) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002Fbigbigwatermalon\u002FC3SQL)\n1. SQLformer: Deep Auto-Regressive Query Graph Generation for Text-to-SQL Translation.\n\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv'2023-purple\"> [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-grey\">](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.18376) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-grey\">](https:\u002F\u002Fgithub.com\u002FAdrianBZG\u002FSQLformer)\n\n## 📊 Text-to-SQL 基准测试\n我们创建了基准测试的发展时间线，并标注了相关的重要里程碑。更多详细信息请参阅本章节：[📊 基准测试](chapter\u002FBenchmark.md)\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002FDataset_timeline.svg\"\u002F>\n\u003C\u002Fp>\n\n## 🎯 我们的目标是什么？\n\n* 🎯解决开放式的 Text-to-SQL 问题\n* 🎯开发经济高效的 Text-to-SQL 方法\n* 🎯使 Text-to-SQL 解决方案更加可信\n* 🎯处理含歧义或未明确说明的自然语言查询\n* 🎯自适应训练数据合成\n\n## 📖 我们的综述目录\n您可以在我们的子章节中获取更多信息。我们介绍了相关概念中的代表性论文：\n* [预处理](chapter\u002FPre_Processing.md)\n* [Text-to-SQL 翻译方法](chapter\u002FTranslation_method.md)\n* [后处理](chapter\u002FPost_Processing.md)\n* [基准测试](chapter\u002FBenchmark.md)\n* [评估](chapter\u002FEvaluation.md)\n* [错误分析](chapter\u002FError_Analysis.md)\n\n## 💾 新手实用指南\n\n### 📊 如何获取数据：\n* 我们为您收集了 Text-to-SQL 基准测试的相关信息及下载链接。更多详情请参阅本章节：[基准测试](chapter\u002FBenchmark.md)\n* 基准测试的分析代码位于 `src\u002Fdataset_analysis` 目录下。基准测试分析报告则存放在 `report\u002F` 目录中。\n\n### 🛠️ 如何构建基于 LLM 的 Text-to-SQL 模型：\n\n* Litgpt [仓库链接](https:\u002F\u002Fgithub.com\u002FLightning-AI\u002Flitgpt)\n\n    该仓库提供了超过20个高性能大型语言模型（LLM）的访问权限，并配有全面的预训练、微调和大规模部署指南。它采用从头开始实现的方式，没有复杂的抽象概念，非常适合初学者使用。\n\n* LLaMA-Factory [仓库链接](https:\u002F\u002Fgithub.com\u002Fhiyouga\u002FLLaMA-Factory)\n    统一高效地微调100多种大型语言模型。通过整合多种模型、可扩展的训练资源、先进的算法、实用技巧以及全面的实验监控工具，该框架能够通过优化后的 API 和用户界面实现高效且快速的推理。\n\n* BIRD-SQL 基准测试的微调与上下文学习 [仓库链接](https:\u002F\u002Fgithub.com\u002FAlibabaResearch\u002FDAMO-ConvAI\u002Ftree\u002Fmain\u002Fbird#fine-tuning-ft)\n    \n    BIRD-SQL 基准测试提供了关于微调和上下文学习的教程。\n\n### 🔎 如何评估您的模型：\n\n我们为您整理了 NL2SQL 的评估指标。更多详情请参阅本章节：[评估](chapter\u002FEvaluation.md)\n\n* NLSQL360 [仓库链接](https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL360) \n\n     NL2SQL360 是一个用于对 NL2SQL 解决方案进行细粒度评估的测试平台。该平台整合了现有的 NL2SQL 基准测试、NL2SQL 模型库以及多种评估指标，旨在提供一个直观且易于使用的平台，以支持标准和定制化的性能评估。 \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEX-red\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEM-green\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVES-blue\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FQVT-orange\">\n\n* Test-suite-sql-eval [仓库链接](https:\u002F\u002Fgithub.com\u002Ftaoyds\u002Ftest-suite-sql-eval)\n\n    该仓库包含针对11个 text-to-SQL 任务的测试套件评估指标。目前，它已成为 [Spider](https:\u002F\u002Fyale-lily.github.io\u002Fspider)、[SParC](https:\u002F\u002Fyale-lily.github.io\u002Fsparc) 和 [CoSQL](https:\u002F\u002Fyale-lily.github.io\u002Fcosql) 的官方评估指标，并且也适用于 Academic、ATIS、Advising、Geography、IMDB、Restaurants、Scholar 和 Yelp 数据集（基于 [Catherine 和 Jonathan](https:\u002F\u002Fgithub.com\u002Fjkkummerfeld\u002Ftext2sql-data) 的出色工作）。 \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEX-red\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEM-green\">\n\n* BIRD-SQL 官方 [仓库链接](https:\u002F\u002Fgithub.com\u002FAlibabaResearch\u002FDAMO-ConvAI\u002Ftree\u002Fmain\u002Fbird#evaluation)\n\n    这是 [BIRD-SQL](https:\u002F\u002Fbird-bench.github.io\u002F) 的官方工具。它是首个提出 VES 指标并提供官方测试套件的工具。 \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEX-red\"> \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FVES-blue\">\n\n\n### 🗺️ 路线图与决策流程\n\n您可以从路线图和决策流程中获得一些启发。\n\u003Cp align=\"center\">\n\u003Cimg width=\"800\" src=\".\u002Fassets\u002FNL2SQL_Guidance.svg\"\u002F>\n\u003C\u002Fp>\n\n## 📱 Text-to-SQL 相关应用：\n\n* AI for Database：面向数据库的代理式 AI 产品——可连接任何数据库（PostgreSQL、MySQL、MongoDB 等），并用通俗英语与其交互。具备自动刷新的智能仪表板、自然语言查询以及基于数据库变化触发的自动化工作流功能。 [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb Link-98f\">](https:\u002F\u002Faifordatabase.com)\n* Chat2DB：基于 AI 的数据库工具和 SQL 客户端，是最受欢迎的 GUI 客户端之一，支持 MySQL、Oracle、PostgreSQL、DB2、SQL Server、SQLite、H2、ClickHouse 等多种数据库。 [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRepositor Link-grey\">](https:\u002F\u002Fgithub.com\u002FcodePhiliaX\u002FChat2DB) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb Link-98f\">](https:\u002F\u002Fchat2db-ai.com\u002Fzh-CN)\n* DB-GPT：基于 AWEL（代理工作流表达语言）和代理的原生 AI 数据应用开发框架。 [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRepositor Link-grey\">](https:\u002F\u002Fgithub.com\u002Feosphoros-ai\u002FDB-GPT) \n* Postgres.new：带有 AI 辅助功能的浏览器内 Postgres 沙盒环境。 [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRepositor Link-grey\">](https:\u002F\u002Fgithub.com\u002Fsupabase-community\u002Fpostgres-new\u002Ftree\u002Fmain) [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb Link-98f\">](https:\u002F\u002Fpostgres.new\u002F)\n* QueryGPT – 使用生成式 AI 将自然语言转换为 SQL。 [\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWeb Link-98f\">](https:\u002F\u002Fwww.uber.com\u002Fen-JP\u002Fblog\u002Fquery-gpt\u002F)\n\n## 📮 联系我们\n如果您认为我们遗漏了某些有趣的工作，请随时与我们联系。\n\n📧 xliu371[at]connect.hkust-gz.edu.cn","# NL2SQL_Handbook 快速上手指南\n\nNL2SQL_Handbook 是一个专注于大模型时代 Text-to-SQL（自然语言转 SQL）技术的综合资源库。它收录了最新的综述论文、教程幻灯片、基准测试以及前沿研究论文列表，旨在帮助开发者和研究人员了解该领域的现状、挑战及未来方向。\n\n> **注意**：本项目主要作为**知识库、文献索引和教程资源**，而非一个直接部署的 Python 软件包。以下指南将指导您如何获取资源并查阅核心内容。\n\n## 环境准备\n\n本项目主要为文档和代码示例集合，对系统环境要求较低，只需具备基础的代码浏览和文档阅读环境。\n\n*   **操作系统**：Linux, macOS, 或 Windows\n*   **前置依赖**：\n    *   **Git**：用于克隆仓库。\n    *   **Markdown 阅读器**：推荐使用 VS Code、Typora 或 GitHub 网页版直接查看 `.md` 文件。\n    *   **PDF 阅读器**：用于查看仓库中提供的学术报告幻灯片（Slides）。\n    *   **Python (可选)**：如果您计划运行仓库中链接的具体子项目代码（如 `NL2SQL360`），建议安装 Python 3.8+ 及相关深度学习框架（PyTorch\u002FTensorFlow），具体依赖需参考对应子项目的说明。\n\n## 安装步骤\n\n由于本项目是资源汇编，无需通过 `pip` 安装，直接克隆仓库即可使用。\n\n### 1. 克隆仓库\n\n打开终端，执行以下命令获取最新资源：\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL_Handbook.git\n```\n\n> **国内加速建议**：如果访问 GitHub 速度较慢，可使用国内镜像源（如 Gitee 镜像，若有）或通过代理加速克隆：\n> ```bash\n> git clone https:\u002F\u002Fgitee.com\u002Fmirror\u002FNL2SQL_Handbook.git  # 示例地址，请以实际存在的镜像为准\n> # 或者使用 gitclone.com 加速\n> git clone https:\u002F\u002Fgitclone.com\u002Fgithub.com\u002FHKUSTDial\u002FNL2SQL_Handbook.git\n> ```\n\n### 2. 进入目录\n\n```bash\ncd NL2SQL_Handbook\n```\n\n### 3. 查看核心资源\n\n克隆完成后，您可以直接在本地浏览以下核心内容：\n\n*   **综述与教程**：查看根目录下的 `README.md` 获取论文列表和链接。\n*   **详细技术章节**：进入 `chapter\u002F` 目录查看模块化技术详解。\n*   **演示文稿**：进入 `slides\u002F` 目录下载 PDF 格式的学术会议报告。\n\n## 基本使用\n\n本手册的核心价值在于其结构化的知识导航。以下是三种最常用的使用方式：\n\n### 1. 查阅模块化技术细节\n\n项目将 Text-to-SQL 流程拆分为三个核心模块，您可以直接阅读对应的 Markdown 文件获取技术实现思路：\n\n*   **预处理 (Pre-processing)**：增强模型输入，解决歧义。\n    ```bash\n    cat chapter\u002FPre_Processing.md\n    ```\n*   **翻译方法 (Translation Methods)**：核心的 NL 到 SQL 转换算法。\n    ```bash\n    cat chapter\u002FTranslation_method.md\n    ```\n*   **后处理 (Post-processing)**：优化生成的 SQL 查询，确保准确性。\n    ```bash\n    cat chapter\u002FPost_Processing.md\n    ```\n\n### 2. 获取最新论文与代码\n\n在 `README.md` 的 **\"Text-to-SQL Paper List\"** 部分，列出了最新的顶会论文（如 SIGMOD, VLDB, ICDE 等）。每项通常包含：\n*   **Paper**：论文 PDF 链接。\n*   **Code**：官方代码仓库链接。\n\n**示例**：若想研究最新的 `Alpha-SQL` 方法，请在列表中查找对应条目，点击 **Code** 按钮跳转至其独立仓库进行安装和运行，因为具体算法的实现代码通常托管在独立的子项目中。\n\n### 3. 学习生命周期与演进\n\n通过查看仓库中的图片资源（位于 `assets\u002F` 目录），直观理解 Text-to-SQL 的生命周期和基于大模型的演进阶段：\n\n*   **生命周期图**：`assets\u002Fnl2sql_lifecycle.svg` (涵盖模型、数据、评估、错误分析)\n*   **演进路线图**：`assets\u002FThe Evolution of NL2SQL Solutions from the Perspective of Language Models.svg`\n\n---\n*提示：如需复现具体的 SOTA 模型，请根据手册中提供的链接前往对应的子项目仓库（例如 `NL2SQL360` 或 `Alpha-SQL`），那里会有详细的 `requirements.txt` 和训练脚本。*","某电商公司的数据团队正试图构建一个让运营人员通过自然语言直接查询销售数据库的系统，但在技术选型和错误优化上陷入了瓶颈。\n\n### 没有 NL2SQL_Handbook 时\n- **技术选型迷茫**：面对海量的 Text-to-SQL 论文，团队难以分辨哪些是仅适用于学术基准的“刷分”模型，哪些真正适合处理复杂的商业嵌套查询。\n- **错误排查无门**：当用户提问“上周复购率最高的品类”生成错误 SQL 时，开发人员缺乏系统的错误分类指南，只能凭经验盲目调试 Prompt。\n- **数据合成困难**：由于缺乏高质量的领域训练数据，团队不知道如何利用最新的数据合成技术来弥补特定业务场景下的样本缺失。\n- **评估标准单一**：仅依赖执行准确率（Execution Accuracy）评估模型，忽略了语义匹配度，导致上线后频繁出现“结果对但逻辑错”的隐蔽风险。\n\n### 使用 NL2SQL_Handbook 后\n- **精准锁定方案**：借助手册中对 LLM 时代技术演进的梳理，团队快速定位到适合处理复杂 Schema 映射的最新架构，大幅缩短了调研周期。\n- **系统化纠错**：利用手册提供的多层次错误分析框架，团队迅速识别出是“模式链接”环节出错，并针对性地引入了修正策略。\n- **高效数据增强**：参考手册中关于数据合成的最佳实践，团队成功生成了贴合电商业务的训练集，显著提升了模型对专业术语的理解力。\n- **多维评估体系**：采纳手册推荐的多粒度评估指标，不仅关注执行结果，更监控中间逻辑，提前拦截了潜在的语义偏差。\n\nNL2SQL_Handbook 将分散的前沿研究转化为可落地的工程指南，帮助团队从“盲目试错”转向“科学构建”，极大降低了自然语言查数的落地门槛。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FHKUSTDial_NL2SQL_Handbook_cf111674.png","HKUSTDial","Data Intelligence and Analytics Lab @ HKUST(GZ)","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FHKUSTDial_47c3425a.jpg","This is the source-code homepage of HKUST(GZ)-DIAL. We are glad to share our code.",null,"yuyuluo@hkust-gz.edu.cn","https:\u002F\u002Fluoyuyu.vip\u002F","https:\u002F\u002Fgithub.com\u002FHKUSTDial",[81],{"name":82,"color":83,"percentage":84},"Python","#3572A5",100,1397,85,"2026-04-10T06:59:34","","未说明",{"notes":91,"python":89,"dependencies":92},"该项目主要是一个综述手册（Handbook）和论文列表资源库，而非一个可直接运行的单一软件工具。README 中未提供具体的安装脚本、环境配置或依赖库列表。它包含了多篇学术论文的链接、幻灯片以及部分相关子项目（如 NL2SQL360, Alpha-SQL 等）的代码仓库链接。若需运行其中提及的具体模型或基准测试，请访问对应的子项目仓库查看其特定的环境需求。",[],[35,13],[95,96,97,98,99,100,101,102,103,104,105,106,107,108,109],"llms","nl2sql","nlp","text-to-sql","awesome","nl-to-code","text2sql","ai4db","text-to-code","db","nl-to-sql","awesome-agents","awesome-nl2sql","awesome-text-to-sql","awesome-text2sql","2026-03-27T02:49:30.150509","2026-04-11T00:41:45.011622",[],[]]