[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-Neph0s--awesome-llm-role-playing-with-persona":3,"tool-Neph0s--awesome-llm-role-playing-with-persona":62},[4,18,26,36,46,54],{"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",159636,2,"2026-04-17T23:33:34",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":42,"last_commit_at":43,"category_tags":44,"status":17},8272,"opencode","anomalyco\u002Fopencode","OpenCode 是一款开源的 AI 编程助手（Coding Agent），旨在像一位智能搭档一样融入您的开发流程。它不仅仅是一个代码补全插件，而是一个能够理解项目上下文、自主规划任务并执行复杂编码操作的智能体。无论是生成全新功能、重构现有代码，还是排查难以定位的 Bug，OpenCode 都能通过自然语言交互高效完成，显著减少开发者在重复性劳动和上下文切换上的时间消耗。\n\n这款工具专为软件开发者、工程师及技术研究人员设计，特别适合希望利用大模型能力来提升编码效率、加速原型开发或处理遗留代码维护的专业人群。其核心亮点在于完全开源的架构，这意味着用户可以审查代码逻辑、自定义行为策略，甚至私有化部署以保障数据安全，彻底打破了传统闭源 AI 助手的“黑盒”限制。\n\n在技术体验上，OpenCode 提供了灵活的终端界面（Terminal UI）和正在测试中的桌面应用程序，支持 macOS、Windows 及 Linux 全平台。它兼容多种包管理工具，安装便捷，并能无缝集成到现有的开发环境中。无论您是追求极致控制权的资深极客，还是渴望提升产出的独立开发者，OpenCode 都提供了一个透明、可信",144296,1,"2026-04-16T14:50:03",[13,45],"插件",{"id":47,"name":48,"github_repo":49,"description_zh":50,"stars":51,"difficulty_score":32,"last_commit_at":52,"category_tags":53,"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":55,"name":56,"github_repo":57,"description_zh":58,"stars":59,"difficulty_score":32,"last_commit_at":60,"category_tags":61,"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",[45,13,15,14],{"id":63,"github_repo":64,"name":65,"description_en":66,"description_zh":67,"ai_summary_zh":67,"readme_en":68,"readme_zh":69,"quickstart_zh":70,"use_case_zh":71,"hero_image_url":72,"owner_login":73,"owner_name":74,"owner_avatar_url":75,"owner_bio":76,"owner_company":77,"owner_location":78,"owner_email":79,"owner_twitter":79,"owner_website":79,"owner_url":80,"languages":79,"stars":81,"forks":82,"last_commit_at":83,"license":79,"difficulty_score":42,"env_os":84,"env_gpu":85,"env_ram":85,"env_deps":86,"category_tags":89,"github_topics":90,"view_count":32,"oss_zip_url":79,"oss_zip_packed_at":79,"status":17,"created_at":108,"updated_at":109,"faqs":110,"releases":111},8893,"Neph0s\u002Fawesome-llm-role-playing-with-persona","awesome-llm-role-playing-with-persona","Awesome-llm-role-playing-with-persona: a curated list of resources for large language models for role-playing with assigned personas","awesome-llm-role-playing-with-persona 是一个精心整理的开源资源库，专注于大语言模型在“角色扮演”领域的应用与研究。它汇集了让 AI 扮演虚构人物、历史名人或特定社会角色的前沿论文、数据集及技术指南，旨在解决大模型在模拟特定人设时容易出现的性格不一致、记忆缺失或缺乏深度个性化等难题。\n\n该资源库不仅涵盖了基础的角色扮演技术，还深入探讨了多智能体协作、拟人化认知（如性格特质、社交智力）以及游戏交互等进阶方向。其独特亮点在于收录了团队最新发表于 TMLR 的综述文章《From Persona to Personalization》，系统梳理了从静态人设设定到动态个性化生成的技术演进路径，为理解该领域提供了权威视角。\n\n无论是从事大模型应用开发的工程师、探索人机交互边界的研究人员，还是希望构建沉浸式对话体验的产品设计师，都能从中找到极具价值的参考素材。通过借鉴其中的方法论与基准测试，用户可以更高效地开发出拥有鲜明个性、能够长期维持角色一致性的智能代理，推动 AI 从单纯的工具向具备情感与个性的伙伴转变。","# Papers and Resources about Role-Playing Language Models\n\n[![Awesome](https:\u002F\u002Fcdn.rawgit.com\u002Fsindresorhus\u002Fawesome\u002Fd7305f38d29fed78fa85652e3a63e154dd8e8829\u002Fmedia\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FNeph0s\u002Fawesome-llm-role-playing-with-persona) \n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-yellow.svg)](https:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT) \n![Last Commit](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flast-commit\u002FNeph0s\u002Fawesome-llm-role-playing-with-persona.svg)\n\n---\n\nThis repo primarily focuses on **character role-playing**, such as fictional characters, celebrities, and historical figures. While role-playing language agents are related to many research topics, such as multi-agent systems and long-context models, we cannot guarantee the inclusion of papers in these areas. \n\n*Here are some other paper lists you might be interested in:*\n\n🤖 ![Last Commit](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flast-commit\u002Fzjunlp\u002FLLMAgentPapers.svg)**[LLM Agent Papers](https:\u002F\u002Fgithub.com\u002Fzjunlp\u002FLLMAgentPapers):**  Must-read Papers on Large Language Model Agents.\n\n## 🔔 News\n \n- **2024-10-09 🎉 Our survey paper\"*From Persona to Personalization:A Survey on Role-Playing Language Agents\"* [[arxiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.18231)] has been accepted to TMLR!**\n\n- **2024-06-28 We reorganized this repo, putting more focus on role-playing agents.**\n\n- **2024-04-30 We are super excited to announce our survey paper: \"*From Persona to Personalization:A Survey on Role-Playing Language Agents\"* [[arxiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.18231)]**\n\n- **2024-04-17 We are looking for collaborators interested in research on this topic. Contact us via email: xtwang21@m.fudan.edu.cn**\n\n- **2023-10-30 We create this repository to maintain a paper list on *LLMs Role-Playing with Assigned Personas* .**\n\n## 📜Content\n\n- [LLM Role-Playing\u002FPersona Papers and Resources](#llm-role-playingpersona-papers-and-resources)\n  - [🔔 News](#-news)\n  - [📜Content](#content)\n    - [🧙 Role-Playing](#-role-playing)\n      - [🧙 Characters](#-characters)\n      - [👨‍👩‍👧‍👦 Demographics](#️-demographics)\n      - [🤖 Personalization](#-personalization)\n    - [👥 Multi Agents](#-multi-agents)\n    - [🤖 GUI Agents for Games](#-gui-agents-for-games)\n    - [🧠 Anthropomorphic Cognition](#-anthropomorphic-cognition)\n      - [🎭 Personality Traits](#-personality-traits)\n      - [👥 Social Intelligence & Theory-of-Mind](#-social-intelligence--theory-of-mind) \n      - [🧠 Others](#-others-1)\n    - [🌐 Others](#-others-2)\n      - [📂 Resources](#-resources)\n      - [📰 Blogs](#-blogs)\n      - [📄 Papers](#-papers)\n  - [💡  Contribution](#--contribution)\n    - [Contributing to this repo](#contributing-to-this-repo)\n    - [Contributors](#contributors)\n\n\n\n### 🧙 Role-Playing\n\n#### 🧙 Characters\n\n*Quick Intro to Role-play LLMs*\n\n-  **From Persona to Personalization: A Survey on Role-Playing Language Agents.**\n   \n   *Jiangjie Chen, Xintao Wang, Rui Xu, Siyu Yuan, Yikai Zhang, Wei Shi, Jian Xie, Shuang Li, Ruihan Yang, Tinghui Zhu, Aili Chen, Nianqi Li, Lida Chen, Caiyu Hu, Siye Wu, Scott Ren, Ziquan Fu, Yanghua Xiao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.18231)], TMLR 2024.04\n\n-  **CoSER: Coordinating LLM-Based Persona Simulation of Established Roles**\n\n   *Xintao Wang, Heng Wang, Yifei Zhang, Xinfeng Yuan, Rui Xu, Jen-tse Huang, Siyu Yuan, Haoran Guo, Jiangjie Chen, Wei Wang, Yanghua Xiao, Shuchang Zhou.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09082)], ICML 2025\n\n-  **RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language Models.**\n\n   *Zekun Moore Wang, Zhongyuan Peng, Haoran Que, Jiaheng Liu, Wangchunshu Zhou, Yuhan Wu, Hongcheng Guo, Ruitong Gan, Zehao Ni, Man Zhang, Zhaoxiang Zhang, Wanli Ouyang, Ke Xu, Wenhu Chen, Jie Fu, Junran Peng.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.00746)], Findings of ACL 2024\n\n-  **ChatHaruhi: Reviving Anime Character in Reality via Large Language Model.**\n\n   *Cheng Li, Ziang Leng, Chenxi Yan, Junyi Shen, Hao Wang, Weishi MI, Yaying Fei, Xiaoyang Feng, Song Yan, HaoSheng Wang, Linkang Zhan, Yaokai Jia, Pingyu Wu, Haozhen Sun.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.09597)], 2023.8\n\n-  **Character-LLM: A Trainable Agent for Role-Playing.**\n\n   *Yunfan Shao, Linyang Li, Junqi Dai, Xipeng Qiu.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.10158)], EMNLP 2023\n\n-  **CharacterEval: A Chinese Benchmark for Role-Playing Conversational Agent Evaluation.**\n   *Quan Tu, Shilong Fan, Zihang Tian, Rui Yan.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.01275)], ACL 2024\n\n-  **InCharacter: Evaluating Personality Fidelity in Role-Playing Agents through Psychological Interviews.**\n\n   *Xintao Wang, Yunze Xiao, Jen-tse Huang, Siyu Yuan, Rui Xu, Haoran Guo, Quan Tu, Yaying Fei, Ziang Leng, Wei Wang, Jiangjie Chen, Cheng Li, Yanghua Xiao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.17976)], ACL 2024\n\n- **BookWorld: From Novels to Interactive Agent Societies for Creative Story Generation**\n\n   *Yiting Ran, Xintao Wang, Tian Qiu, Jiaqing Liang, Yanghua Xiao, Deqing Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.14538)], ACL 2025\n\n- **Character is Destiny: Can Large Language Models Simulate Persona-Driven Decisions in Role-Playing?**\n\n   *Rui Xu, Xintao Wang, Jiangjie Chen, Siyu Yuan, Xinfeng Yuan, Jiaqing Liang, Zulong Chen, Xiaoqing Dong, Yanghua Xiao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.12138)], Findings of EMNLP 2025\n\n- **CharacterBench: Benchmarking Character Customization of Large Language Models**\n   \n   *Jinfeng Zhou, Yongkang Huang, Bosi Wen, Guanqun Bi, Yuxuan Chen, Pei Ke, Zhuang Chen, Xiyao Xiao, Libiao Peng, Kuntian Tang, Rongsheng Zhang, Le Zhang, Tangjie Lv, Zhipeng Hu, Hongning Wang, Minlie Huang* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.11912)], AAAI 2025\n\n-  **OmniCharacter: Towards Immersive Role-Playing Agents with Seamless Speech-Language Personality Interaction**\n\n   *Haonan Zhang, Run Luo, Xiong Liu, Yuchuan Wu, Ting-En Lin, Pengpeng Zeng, Qiang Qu, Feiteng Fang, Min Yang, Lianli Gao, Jingkuan Song, Fei Huang, Yongbin Li.* [[abs](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2505.20277)], ACL 2025\n\n- **Persona vectors: Monitoring and controlling character traits in language models**\n   *Runjin Chen, Andy Arditi, Henry Sleight, Owain Evans, Jack Lindsey.* [[abs](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Fpersona-vectors)], Anthropic Blog, 2025.08\n\n\n- **CPO: Addressing Reward Ambiguity in Role-playing Dialogue via Comparative Policy Optimization**\n   *Xinge Ye, Rui Wang, Yuchuan Wu, Victor Ma, Feiteng Fang, Fei Huang, Yongbin Li.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.09074)], Tongyi Lab, 2025.08\n\n- **CogDual: Enhancing Dual Cognition of LLMs via Reinforcement Learningwith Implicit Rule-Based Rewards**\n   *Cheng Liu, Yifei Lu, Fanghua Ye, Jian Li, Xingyu Chen, Feiliang Ren, Zhaopeng Tu, Xiaolong Li.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.17147)], 2025.07\n\n- **The Assistant Axis: Situating and Stabilizing the Default Persona of Language Models**\n   *Christina Lu, Jack Gallagher, Jonathan Michala, Kyle Fish, Jack Lindsey.* [[abs](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2601.10387)], 2026.01\n\n- **Emotion Concepts and their Function in a Large Language Model**\n   *Nicholas Sofroniew, Isaac Kauvar, William Saunders, Runjin Chen, Tom Henighan, Sasha Hydrie, Craig Citro, Adam Pearce, Julius Tarng, Wes Gurnee, Joshua Batson, Sam Zimmerman, Kelley Rivoire, Kyle Fish, Chris Olah, Jack Lindsey.* [[web](https:\u002F\u002Ftransformer-circuits.pub\u002F2026\u002Femotions\u002Findex.html)], 2026.04\n\n\n*All*\n*Established Characters*\n\n- **HER: Human-like Reasoning and Reinforcement Learning for LLM Role-playing**\n   *Chengyu Du, Xintao Wang, Aili Chen, Weiyuan Li, Rui Xu, Junteng Liu, Zishan Huang, Rong Tian, Zijun Sun, Yuhao Li, Liheng Feng, Deming Ding, Pengyu Zhao, Yanghua Xiao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.21459)], Findings of ACL 2026\n\n- **HumanLLM: Benchmarking and Improving LLM Anthropomorphism via Human Cognitive Patterns**\n   *Xintao Wang, Jian Yang, Weiyuan Li, Rui Xie, Jen-tse Huang, Jun Gao, Shuai Huang, Yueping Kang, Liyuan Gou, Hongwei Feng, Yanghua Xiao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.10198)], ACL 2026\n\n\n- **MOA: Multi-Objective Alignment for Role-Playing Agents**\n   *Chonghua Liao, Ke Wang, Yuchuan Wu, Fei Huang, Yongbin Li.* [[abs](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2512.09756)],  2025.12\n\n- **Too Good to be Bad: On the Failure of LLMs to Role-Play Villains**\n   *Zihao Yi, Qingxuan Jiang, Ruotian Ma, Xingyu Chen, Qu Yang, Mengru Wang, Fanghua Ye, Ying Shen, Zhaopeng Tu, Xiaolong Li, Linus.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.04962)],  2025.11\n\n- **CPO: Addressing Reward Ambiguity in Role-playing Dialogue via Comparative Policy Optimization**\n   *Xinge Ye, Rui Wang, Yuchuan Wu, Victor Ma, Feiteng Fang, Fei Huang, Yongbin Li.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.09074)], Tongyi Lab, 2025.08\n\n- **RMTBench: Benchmarking LLMs Through Multi-Turn User-Centric Role-Playing**\n   *Hao Xiang, Tianyi Tang, Yang Su, Bowen Yu, An Yang, Fei Huang, Yichang Zhang, Yaojie Lu, Hongyu Lin, Xianpei Han, Jingren Zhou, Junyang Lin, Le Sun.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.20352)], 2025.08\n\n- **CogDual: Enhancing Dual Cognition of LLMs via Reinforcement Learningwith Implicit Rule-Based Rewards**\n   *Cheng Liu, Yifei Lu, Fanghua Ye, Jian Li, Xingyu Chen, Feiliang Ren, Zhaopeng Tu, Xiaolong Li.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.17147)], 2025.07\n\n- **Spotting Out-of-Character Behavior: Atomic-Level Evaluation of Persona Fidelity in Open-Ended Generation**\n   *Jisu Shin, Juhyun Oh, Eunsu Kim, Hoyun Song, Alice Oh.* [[abs](https:\u002F\u002Faclanthology.org\u002F2025.findings-acl.1349\u002F)], Findings of ACL 2025\n\n- **Crab: A Novel Configurable Role-Playing LLM with Assessing Benchmark**\n   *Kai He, Yucheng Huang, Wenqing Wang, Delong Ran, Dongming Sheng, Junxuan Huang, Qika Lin, Jiaxing Xu, Wenqiang Liu, Mengling Feng.* [[abs](https:\u002F\u002Faclanthology.org\u002F2025.acl-long.731\u002F)], ACL 2025\n \n- **Test-Time-Matching: Decouple Personality, Memory, and Linguistic Style in LLM-based Role-Playing Language Agent**\n   \n   *Xiaoyu Zhan, Xinyu Fu, Hao Sun, Yuanqi Li, Jie Guo, Yanwen Guo.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.16799)], 2025.07\n\n- **Thinking in Character: Advancing Role-Playing Agents with Role-Aware Reasoning**\n\n   *Yihong Tang, Kehai Chen, Muyun Yang, Zhengyu Niu, Jing Li, Tiejun Zhao, Min Zhang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.01748)], 2025.06\n\n\n- **BookWorld: From Novels to Interactive Agent Societies for Creative Story Generation**\n\n   *Yiting Ran, Xintao Wang, Tian Qiu, Jiaqing Liang, Yanghua Xiao, Deqing Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.14538)], ACL 2025\n   \n-  **OmniCharacter: Towards Immersive Role-Playing Agents with Seamless Speech-Language Personality Interaction**\n\n   *Haonan Zhang, Run Luo, Xiong Liu, Yuchuan Wu, Ting-En Lin, Pengpeng Zeng, Qiang Qu, Feiteng Fang, Min Yang, Lianli Gao, Jingkuan Song, Fei Huang, Yongbin Li.* [[abs](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2505.20277)], ACL 2025\n\n- **PsyMem: Fine-grained psychological alignment and Explicit Memory Control for Advanced Role-Playing LLMs**\n   \n   *Xilong Cheng, Yunxiao Qin, Yuting Tan, Zhengnan Li, Ye Wang, Hongjiang Xiao, Yuan Zhang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.12814)], 2025.05\n\n- **RAIDEN-R1: Improving Role-awareness of LLMs via GRPO with Verifiable Reward**\n   *Zongsheng Wang, Kaili Sun, Bowen Wu, Qun Yu, Ying Li, Baoxun Wang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.10218)], 2025.05\n\n- **ChARM: Character-based Act-adaptive Reward Modeling for Advanced Role-Playing Language Agents**\n   *Feiteng Fang, Ting-En Lin, Yuchuan Wu, Xiong Liu, Xiang Huang, Dingwei Chen, Jing Ye, Haonan Zhang, Liang Zhu, Hamid Alinejad-Rokny, Min Yang, Fei Huang, Yongbin Li.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.23923)], 2025.05\n\n- **Codifying Character Logic in Role-Playing**\n   *Letian Peng, Jingbo Shang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.07705)], 2025.05\n\n- **When Harry Meets Superman: The Role of The Interlocutor in Persona-Based Dialogue Generation**\n   *Daniela Occhipinti, Marco Guerini, Malvina Nissim.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.24613)], 2025.05\n\n-  **CoSER: Coordinating LLM-Based Persona Simulation of Established Roles**\n\n   *Xintao Wang, Heng Wang, Yifei Zhang, Xinfeng Yuan, Rui Xu, Jen-tse Huang, Siyu Yuan, Haoran Guo, Jiangjie Chen, Wei Wang, Yanghua Xiao, Shuchang Zhou.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09082)], ICML 2025\n   \n- **Towards a Design Guideline for RPA Evaluation: A Survey of Large Language Model-Based Role-Playing Agents**\n   *Chaoran Chen, Bingsheng Yao, Ruishi Zou, Wenyue Hua, Weimin Lyu, Yanfang Ye, Toby Jia-Jun Li, Dakuo Wang* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13012)], 2025.02\n\n- **RPGBENCH: Evaluating Large Language Models as Role-Playing Game Engines**\n   *Pengfei Yu, Dongming Shen, Silin Meng, Jaewon Lee, Weisu Yin, Andrea Yaoyun Cui, Zhenlin Xu, Yi Zhu, Xingjian Shi, Mu Li, Alex Smola*[[abs](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2502.00595)], 2025.02\n\n- **Reasoning Does Not Necessarily Improve Role-Playing Ability**\n   *Xiachong Feng, Longxu Dou, Lingpeng Kong.* [[abs]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.16940), 2025.02\n\n- **CharacterBox: Evaluating the Role-Playing Capabilities of LLMs in Text-Based Virtual Worlds**\n   *Lei Wang, Jianxun Lian, Yi Huang, Yanqi Dai, Haoxuan Li, Xu Chen, Xing Xie, Ji-Rong Wen.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.05631)], 2024.12\n\n- **Revealing and Mitigating the Challenge of Detecting Character Knowledge Errors in LLM Role-Playing**\n   *Wenyuan Zhang, Shuaiyi Nie, Jiawei Sheng, Zefeng Zhang, Xinghua Zhang, Yongquan He, Tingwen Liu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11726)], 2024.09\n\n-  **Mitigating Hallucination in Fictional Character Role-Play**\n\n   *Nafis Sadeq, Zhouhang Xie, Byungkyu Kang, Prarit Lamba, Xiang Gao, Julian McAuley.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.17260)] [[dataset](https:\u002F\u002Fgithub.com\u002FNafisSadeq\u002Frolefact)], Findings of EMNLP 2024\n\n-  **Quantifying and Optimizing Global Faithfulness in Persona-driven Role-playing**\n\n   *Letian Peng, Jingbo Shang.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.07726)] [[dataset](https:\u002F\u002Fgithub.com\u002FKomeijiForce\u002FActive_Passive_Constraint_Koishiday_2024)], NeurIPS 2024\n\n-  **CharacterGLM: Customizing Chinese Conversational AI Characters with Large Language Models.**\n   *Jinfeng Zhou, Zhuang Chen, Dazhen Wan, Bosi Wen, Yi Song, Jifan Yu, Yongkang Huang, Libiao Peng, Jiaming Yang, Xiyao Xiao, Sahand Sabour, Xiaohan Zhang, Wenjing Hou, Yijia Zhang, Yuxiao Dong, Jie Tang, Minlie Huang.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.16832)], EMNLP 2024, Industrial Track\n\n-  **PIPPA: A Partially Synthetic Conversational Dataset**\n   *Tear Gosling, Alpin Dale, Yinhe Zheng.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.05884)], 2023.08\n\n- **Character is Destiny: Can Large Language Models Simulate Persona-Driven Decisions in Role-Playing?**\n\n   *Rui Xu, Xintao Wang, Jiangjie Chen, Siyu Yuan, Xinfeng Yuan, Jiaqing Liang, Zulong Chen, Xiaoqing Dong, Yanghua Xiao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.12138)], Findings of EMNLP 2025\n\n\n-  **Large Language Models are Superpositions of All Characters: Attaining Arbitrary Role-play via Self-Alignment**\n   *Keming Lu, Bowen Yu, Chang Zhou, Jingren Zhou.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.12474)], 2024.01\n\n-  **Large Language Models Meet Harry Potter: A Bilingual Dataset for Aligning Dialogue Agents with Characters.**\n   *Nuo Chen, Yan Wang, Haiyun Jiang, Deng Cai, Yuhan Li, Ziyang Chen, Longyue Wang, Jia Li.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.06869)], 2022.11\n\n- **Evaluating Character Understanding of Large Language Models via Character Profiling from Fictional Works**\n   *Xinfeng Yuan, Siyu Yuan, Yuhan Cui, Tianhe Lin, Xintao Wang, Rui Xu, Jiangjie Chen, Deqing Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.12726)], 2024.04\n\n\n- **From Role-Play to Drama-Interaction: An LLM Solution**\n   *Weiqi Wu, Hongqiu Wu, Lai Jiang, Xingyuan Liu, Jiale Hong, Hai Zhao, Min Zhang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.14231)], Findings of ACL 2024\n\n- **TimeChara: Evaluating Point-in-Time Character Hallucination of Role-Playing Large Language Models**\n   *Jaewoo Ahn, Taehyun Lee, Junyoung Lim, Jin-Hwa Kim, Sangdoo Yun, Hwaran Lee, Gunhee Kim.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.18027)], Findings of ACL 2024\n\n\n-  **\"Let Your Characters Tell Their Story\": A Dataset for Character-Centric Narrative Understanding**\n   *Faeze Brahman, Meng Huang, Oyvind Tafjord, Chao Zhao, Mrinmaya Sachan, Snigdha Chaturvedi.*, [[abs](https:\u002F\u002Faclanthology.org\u002F2021.findings-emnlp.150\u002F)], EMNLP findings 2021\n\n\n-  **CharacterEval: A Chinese Benchmark for Role-Playing Conversational Agent Evaluation.**\n   *Quan Tu, Shilong Fan, Zihang Tian, Rui Yan.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.01275)], ACL 2024\n\n\n-  **Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent**\n   *Xiaoyan Yu, Tongxu Luo, Yifan Wei, Fangyu Lei, Yiming Huang, Hao Peng, Liehuang Zhu.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.13717)], 2024.02\n\n-  **RoleEval: A Bilingual Role Evaluation Benchmark for Large Language Models**\n   *Tianhao Shen, Sun Li, Quan Tu, Deyi Xiong.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.16132)], 2023.12\n\n-  **ERABAL: Enhancing Role-Playing Agents through Boundary-Aware Learning**\n   *Yihong Tang, Jiao Ou, Che Liu, Fuzheng Zhang, Di Zhang, Kun Gai.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.14710)], 2024.09\n\n- **SocialBench: Sociality Evaluation of Role-Playing Conversational Agents**\n   *Hongzhan Chen, Hehong Chen, Ming Yan, Wenshen Xu, Xing Gao, Weizhou Shen, Xiaojun Quan, Chenliang Li, Ji Zhang, Fei Huang, Jingren Zhou.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.13679)], 2024.03\n\n- **Capturing Minds, Not Just Words: Enhancing Role-Playing Language Models with Personality-Indicative Data**\n   *Yiting Ran, Xintao Wang, Rui Xu, Xinfeng Yuan, Jiaqing Liang, Yanghua Xiao, Deqing Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.18921)], Findings of EMNLP 2024\n\n- **MMRole: A Comprehensive Framework for Developing and Evaluating Multimodal Role-Playing Agents**\n   *Yanqi Dai, Huanran Hu, Lei Wang, Shengjie Jin, Xu Chen, Zhiwu Lu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.04203)], 2024.08\n\n- **Revealing the Challenge of Detecting Character Knowledge Errors in LLM Role-Playing**\n   *Wenyuan Zhang, Jiawei Sheng, Shuaiyi Nie, Zefeng Zhang, Xinghua Zhang, Yongquan He, Tingwen Liu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11726)], 2024.09\n\n- **BEYOND DIALOGUE: A Profile-Dialogue Alignment Framework Towards General Role-Playing Language Model**\n   *Yeyong Yu, Runsheng Yu, Haojie Wei, Zhanqiu Zhang, Quan Qian.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.10903)], 2024.08\n\n- **PRODIGy: a PROfile-based DIalogue Generation dataset**\n  *Daniela Occhipinti, Serra Sinem Tekiroglu, Marco Guerini* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.05195)] [[dataset](https:\u002F\u002Fgithub.com\u002FLanD-FBK\u002Fprodigy-dataset)], NAACL Findings 2024\n\n- **CharacterBench: Benchmarking Character Customization of Large Language Models**\n   *Jinfeng Zhou, Yongkang Huang, Bosi Wen, Guanqun Bi, Yuxuan Chen, Pei Ke, Zhuang Chen, Xiyao Xiao, Libiao Peng, Kuntian Tang, Rongsheng Zhang, Le Zhang, Tangjie Lv, Zhipeng Hu, Hongning Wang, Minlie Huang* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.11912)], AAAI 2025\n\n*Synthetic Characters*\n\n- **LLM Generated Persona is a Promise with a Catch**\n   *Ang Li, Haozhe Chen, Hongseok Namkoong, Tianyi Peng.* [[abs](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.16527)], 2025.03\n\n- **RoleMRC: A Fine-Grained Composite Benchmark for Role-Playing and Instruction-Following**\n   *Junru Lu, Jiazheng Li, Guodong Shen, Lin Gui, Siyu An, Yulan He, Di Yin, Xing Sun.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.16527)], 2025.02\n\n\n- **OpenCharacter: Training Customizable Role-Playing LLMs with Large-Scale Synthetic Personas**\n   *Xiaoyang Wang, Hongming Zhang, Tao Ge, Wenhao Yu, Dian Yu, Dong Yu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.15427)], 2025.01\n\n- **Scaling Synthetic Data Creation with 1,000,000,000 Personas**\n   *Xin Chan, Xiaoyang Wang, Dian Yu, Haitao Mi, Dong Yu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.20094)], 2024.06\n\n- **Stark: Social Long-Term Multi-Modal Conversation with Persona Commonsense Knowledge**\n   *Young-Jun Lee, Dokyong Lee, Junyoung Youn, Kyeongjin Oh, Byungsoo Ko, Jonghwan Hyeon, Ho-Jin Choi.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.03958)], EMNLP 2024\n\n- **PersonaGym: Evaluating Persona Agents and LLMs**\n   *Vinay Samuel, Henry Peng Zou, Yue Zhou, Shreyas Chaudhari, Ashwin Kalyan, Tanmay Rajpurohit, Ameet Deshpande, Karthik Narasimhan, Vishvak Murahari.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.18416)], 2024.10\n\n#### 👨‍👩‍👧‍👦 Demographics\n\n- **Aligning Large Language Models with Human Opinions through Persona Selection and Value–Belief–Norm Reasoning**\n   *Xuan Long Do, Kenji Kawaguchi, Min-Yen Kan, Nancy Chen.* [[abs](https:\u002F\u002Faclanthology.org\u002F2025.coling-main.172.pdf)], COLING 2025\n\n-  **Expertprompting: Instructing large language models to be distinguished experts**\n\n   *Benfeng Xu , An Yang , Junyang Lin , Quan Wang , Chang Zhou, Yongdong Zhang and Zhendong Mao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.14688)], 2023.5\n\n-  **Toxicity in ChatGPT: Analyzing Persona-assigned Language Models**\n\n   *Ameet Deshpande, Vishvak Murahari, Tanmay Rajpurohit, Ashwin Kalyan, Karthik Narasimhan.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.05335)], 2023.4\n\n-  **Better Zero-Shot Reasoning with Role-Play Prompting**\n   *Aobo Kong, Shiwan Zhao, Hao Chen, Qicheng Li, Yong Qin, Ruiqi Sun, Xin Zhou.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.07702)], 2023.8\n\n-  **Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs**\n   *Shashank Gupta, Vaishnavi Shrivastava, Ameet Deshpande, Ashwin Kalyan, Peter Clark, Ashish Sabharwal, Tushar Khot* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.04892)], 2023.11\n\n-  **Is \"A Helpful Assistant\" the Best Role for Large Language Models? A Systematic Evaluation of Social Roles in System Prompts.**\n   *Mingqian Zheng, Jiaxin Pei, David Jurgens* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.10054)], 2023.11\n\n-  **CoMPosT: Characterizing and Evaluating Caricature in LLM Simulations**\n   *Myra Cheng, Tiziano Piccardi, Diyi Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.11501)], 2023.10\n\n-  **Deciphering Stereotypes in Pre-Trained Language Models**\n   *Weicheng Ma, Henry Scheible, Brian Wang, Goutham Veeramachaneni, Pratim Chowdhary, Alan Sun, Andrew Koulogeorge, Lili Wang, Diyi Yang, Soroush Vosoughi.* [[abs](https:\u002F\u002Faclanthology.org\u002F2023.emnlp-main.697\u002F)], EMNLP Findings 2023\n\n-  **CultureLLM: Incorporating Cultural Differences into Large Language Models**\n   *Cheng Li, Mengzhou Chen, Jindong Wang, Sunayana Sitaram, Xing Xie.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.10946)], 2024.02\n\n- **Aligning Large Language Models with Human Opinions through Persona Selection and Value–Belief–Norm Reasoning**\n   *Xuan Long Do, Kenji Kawaguchi, Min-Yen Kan, Nancy Chen.* [[abs](https:\u002F\u002Faclanthology.org\u002F2025.coling-main.172.pdf)], COling 2025\n\n#### 🤖 Personalization\n\n- **Know Me, Respond to Me: Benchmarking LLMs for Dynamic User Profiling and Personalized Responses at Scale**\n   *Bowen Jiang, Zhuoqun Hao, Young-Min Cho, Bryan Li, Yuan Yuan, Sihao Chen, Lyle Ungar, Camillo J. Taylor, Dan Roth* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.14225)], 2025.04\n\n- **DEEPER Insight into Your User: Directed Persona Refinement for Dynamic Persona Modeling**\n   *Aili Chen, Chengyu Du, Jiangjie Chen, Jinghan Xu, Yikai Zhang, Siyu Yuan, Zulong Chen, Liangyue Li, Yanghua Xiao* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.11078)], 2025.02\n\n- **How Far are LLMs from Being Our Digital Twins? A Benchmark for Persona-Based Behavior Chain Simulation**\n   *Rui Li, Heming Xia, Xinfeng Yuan, Qingxiu Dong, Lei Sha, Wenjie Li, Zhifang Sui* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.14642)], 2025.02\n\n-  **LaMP: When Large Language Models Meet Personalization**\n   *Alireza Salemi, Sheshera Mysore, Michael Bendersky, Hamed Zamani* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.11406)], 2023.04.\n\n- **How Well Can LLMs Echo Us? Evaluating AI Chatbots' Role-Play Ability with ECHO**\n   *Man Tik Ng, Hui Tung Tse, Jen-tse Huang, Jingjing Li, Wenxuan Wang, Michael R. Lyu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.13957)], 2024.04\n\n-  **On the steerability of large language models toward data-driven personas**\n   *Junyi Li, Ninareh Mehrabi, Charith Peris, Palash Goyal, Kai-Wei Chang, Aram Galstyan, Richard Zemel, Rahul Gupta* [[abs](https:\u002F\u002Faclanthology.org\u002F2024.naacl-long.405\u002F)], NAACL 2024.\n\n- **Show, Don't Tell: Aligning Language Models with Demonstrated Feedback**\n   *Omar Shaikh, Michelle Lam, Joey Hejna, Yijia Shao, Michael Bernstein, Diyi Yang* [[abs](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2406.00888)], 2024.06\n\n- **PersLLM: A Personified Training Approach for Large Language Models**\n   *Zheni Zeng, Jiayi Chen, Huimin Chen, Yukun Yan, Yuxuan Chen, Zhenghao Liu, Zhiyuan Liu, Maosong Sun* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.12393)], 2024.07\n\n-  **LiveChat: A Large-Scale Personalized Dialogue Dataset Automatically Constructed from Live Streaming**\n\n   *Jingsheng Gao, Yixin Lian, Ziyi Zhou, Yuzhuo Fu, Baoyuan Wang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.08401)], 2023.6. Tags: streamer persona.\n\n-  **COSPLAY: Concept Set Guided Personalized Dialogue Generation Across Both Party Personas**\n   *Jingsheng Gao, Yixin Lian, Ziyi Zhou, Yuzhuo Fu, Baoyuan Wang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.00872)], SIGIR 2022.\n\n-  **Personalizing Dialogue Agents: I have a dog, do you have pets too?** \n   *Saizheng Zhang, Emily Dinan, Jack Urbanek, Arthur Szlam, Douwe Kiela, Jason Weston.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.07243)], ACL 2018.\n\n-  **MPCHAT: Towards Multimodal Persona-Grounded Conversation**\n   *Jaewoo Ahn, Yeda Song, Sangdoo Yun, Gunhee Kim.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.17388)], ACL 2023. Tags: multimodal persona.\n\n\n-  **Personalized Reward Learning with Interaction-Grounded Learning (IGL)**\n   *Jessica Maghakian, Paul Mineiro, Kishan Panaganti, Mark Rucker, Akanksha Saran, Cheng Tan* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.15823)], 2022.11.\n\n-  **When Large Language Models Meet Personalization: Perspectives of Challenges and Opportunities**\n   *Jin Chen, Zheng Liu, Xu Huang, Chenwang Wu, Qi Liu, Gangwei Jiang, Yuanhao Pu, Yuxuan Lei, Xiaolong Chen, Xingmei Wang, Defu Lian, Enhong Chen* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.16376)], 2023.07, Survey.\n\n-  **Beyond ChatBots: ExploreLLM for Structured Thoughts and Personalized Model Responses**\n   *Xiao Ma, Swaroop Mishra, Ariel Liu, Sophie Su, Jilin Chen, Chinmay Kulkarni, Heng-Tze Cheng, Quoc Le, Ed Chi* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.00763)], 2023.12.\n\n- **PersonalityChat: Conversation Distillation for Personalized Dialog Modeling with Facts and Traits**\n   *Ehsan Lotfi, Maxime De Bruyn, Jeska Buhmann, Walter Daelemans.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.07363)], 2024.01.\n\n- **Beyond Demographics: Aligning Role-playing LLM-based Agents Using Human Belief Networks**\n   *Yun-Shiuan Chuang, Krirk Nirunwiroj, Zach Studdiford, Agam Goyal, Vincent V. Frigo, Sijia Yang, Dhavan Shah, Junjie Hu, Timothy T. Rogers.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.17232)], 2024.06\n\n- **Generative Agent Simulations of 1,000 People**\n   *Joon Sung Park, Carolyn Q. Zou, Aaron Shaw, Benjamin Mako Hill, Carrie Cai, Meredith Ringel Morris, Robb Willer, Percy Liang, Michael S. Bernstein.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.10109)], 2024.11\n\n\n### 👥 Multi Agents\n\n#### 🧙 Characters\n\n-  **Generative Agents: Interactive Simulacra of Human Behavior**\n   *Joon Sung Park, Joseph C. O'Brien, Carrie J. Cai, Meredith Ringel Morris, Percy Liang, Michael S. Bernstein.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.03442)], 2023.04.\n\n-  **Communicative Agents for Software Development**\n   *Chen Qian, Xin Cong, Wei Liu, Cheng Yang, Weize Chen, Yusheng Su, Yufan Dang, Jiahao Li, Juyuan Xu, Dahai Li, Zhiyuan Liu, Maosong Sun.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.07924)], 2023.07.\n\n-  **Corex: Pushing the Boundaries of Complex Reasoning through Multi-Model Collaboration.**\n   *Qiushi Sun, Zhangyue Yin, Xiang Li, Zhiyong Wu, Xipeng Qiu, Lingpeng Kong.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.00280)], 2023.09.\n\n-  **AVALONBENCH: Evaluating LLMs Playing the Game of Avalon.**\n   *Jonathan Light, Min Cai, Sheng Shen, Ziniu Hu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.05036)], 2023.10.\n\n-  **War and Peace (WarAgent): Large Language Model-based Multi-Agent Simulation of World Wars.**\n   *Wenyue Hua, Lizhou Fan, Lingyao Li, Kai Mei, Jianchao Ji, Yingqiang Ge, Libby Hemphill, Yongfeng Zhang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.17227)], 2023.11.\n\n-  **Leveraging Word Guessing Games to Assess the Intelligence of Large Language Models.**\n   *Tian Liang, Zhiwei He, Jen-tse Huang, Wenxuan Wang, Wenxiang Jiao, Rui Wang, Yujiu Yang, Zhaopeng Tu, Shuming Shi, Xing Wang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.20499)], 2023.10.\n\n-  **AntEval: Quantitatively Evaluating Informativeness and Expressiveness of Agent Social Interactions.**\n   *Yuanzhi Liang, Linchao Zhu, Yi Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.06509)], 2024.01.\n\n-  **Exchange-of-Thought: Enhancing Large Language Model Capabilities through Cross-Model Communication**\n   *Zhangyue Yin, Qiushi Sun, Cheng Chang, Qipeng Guo, Junqi Dai, Xuanjing Huang, Xipeng Qiu.* [[abs](https:\u002F\u002Faclanthology.org\u002F2023.emnlp-main.936\u002F)], 2024.01.\n\n-  **Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology View**\n   *Jintian Zhang, Xin Xu, Ningyu Zhang, Ruibo Liu, Bryan Hooi, Shumin Deng.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.02124)], 2023.10\n\n-  **Training socially aligned language models on simulated social interactions**\n   *Ruibo Liu, Ruixin Yang, Chenyan Jia, Ge Zhang, Diyi Yang, Soroush Vosoughi.* [[abs](https:\u002F\u002Fopenreview.net\u002Fforum?id=NddKiWtdUm)], ICLR 2024\n\n-  **How Far Are We on the Decision-Making of LLMs? Evaluating LLMs' Gaming Ability in Multi-Agent Environments**\n   *Jen-tse Huang, Eric John Li, Man Ho Lam, Tian Liang, Wenxuan Wang, Youliang Yuan, Wenxiang Jiao, Xing Wang, Zhaopeng Tu, Michael R. Lyu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.11807)], 2024.03\n\n-  **AgentGroupChat: An Interactive Group Chat Simulacra For Better Eliciting Collective Emergent Behavior**\n   *Zhouhong Gu, Xiaoxuan Zhu, Haoran Guo, Lin Zhang, Yin Cai, Hao Shen, Jiangjie Chen, Zheyu Ye, Yifei Dai, Yan Gao, Yao Hu, Hongwei Feng, Yanghua Xiao* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.13433)], 2024.03\n\n-  **Is this the real life? Is this just fantasy? The Misleading Success of Simulating Social Interactions With LLMs**\n   *Xuhui Zhou, Zhe Su, Tiwalayo Eisape, Hyunwoo Kim, Maarten Sap.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.05020)], 2024.03\n\n- **EvoAgent: Towards Automatic Multi-Agent Generation via Evolutionary Algorithms**\n  *Siyu Yuan, Kaitao Song, Jiangjie Chen, Xu Tan, Dongsheng Li, Deqing Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.14228)], 2024.06\n\n- **HoLLMwood: Unleashing the Creativity of Large Language Models in Screenwriting via Role Playing**\n   *Jing Chen, Xinyu Zhu, Cheng Yang, Chufan Shi, Yadong Xi, Yuxiang Zhang, Junjie Wang, Jiashu Pu, Rongsheng Zhang, Yujiu Yang, Tian Feng* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.11683)], 2024.06\n\n- **MathVC: An LLM-Simulated Multi-Character Virtual Classroom for Mathematics Education**\n  *Murong Yue, Wijdane Mifdal, Yixuan Zhang, Jennifer Suh, Ziyu Yao.* [[abs](https:\u002F\u002Fmurongyue.github.io\u002FMathVC.github.io\u002F)], 2024.4\n\n- **Artificial Leviathan: Exploring Social Evolution of LLM Agents Through the Lens of Hobbesian Social Contract Theory**\n   *Gordon Dai, Weijia Zhang, Jinhan Li, Siqi Yang, Chidera Onochie lbe, Srihas Rao, Arthur Caetano, Misha Sra.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.14373)], 2024.06\n\n\n- **Dialogue Action Tokens: Steering Language Models in Goal-Directed Dialogue with a Multi-Turn Planner**\n   *Kenneth Li, Yiming Wang, Fernanda Viégas, Martin Wattenberg.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.11978)], 2024.06\n\n- **LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Play**\n   *Li-Chun Lu, Shou-Jen Chen, Tsung-Min Pai, Chan-Hung Yu, Hung-yi Lee, Shao-Hua Sun.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.06373)], COLM 2024\n\n- **Persona Inconstancy in Multi-Agent LLM Collaboration: Conformity, Confabulation, and Impersonation**\n   *Razan Baltaji, Babak Hemmatian, Lav R. Varshney.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.03862)], 2024.05\n\n-  **Multi-expert Prompting Improves Reliability, Safety, and Usefulness of Large Language Models**\n\n   *Do Xuan Long, Duong Ngoc Yen, Anh Tuan Luu, Kenji Kawaguchi, Min-Yen Kan, Nancy F. Chen.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.00492)], EMNLP 2024.\n\n- **Project Sid: Many-agent simulations toward AI civilization**\n   *Altera.AL, Andrew Ahn, Nic Becker, Stephanie Carroll, Nico Christie, Manuel Cortes, Arda Demirci, Melissa Du, Frankie Li, Shuying Luo, Peter Y Wang, Mathew Willows, Feitong Yang, Guangyu Robert Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.00114)], 2024.11\n\n- **SocioVerse: A World Model for Social Simulation Powered by LLM Agents and A Pool of 10 Million Real-World Users**\n   *Xinnong Zhang, Jiayu Lin, Xinyi Mou, Shiyue Yang, Xiawei Liu, Libo Sun, Hanjia Lyu, Yihang Yang, Weihong Qi, Yue Chen, Guanying Li, Ling Yan, Yao Hu, Siming Chen, Yu Wang, Xuanjing Huang, Jiebo Luo, Shiping Tang, Libo Wu, Baohua Zhou, Zhongyu Wei.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.10157)], 2025.04\n\n- **Modeling Earth-Scale Human-Like Societies with One Billion Agents**\n   *Haoxiang Guan, Jiyan He, Liyang Fan, Zhenzhen Ren, Shaobin He, Xin Yu, Yuan Chen, Shuxin Zheng, Tie-Yan Liu, Zhen Liu.* [[abs](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2506.12078)], 2025.06\n\n\n\n\n\n   *Do Xuan Long, Duong Ngoc Yen, Anh Tuan Luu, Kenji Kawaguchi, Min-Yen Kan, Nancy F. Chen.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.00492)], EMNLP 2024.\n\n- **Project Sid: Many-agent simulations toward AI civilization**\n   *Altera.AL, Andrew Ahn, Nic Becker, Stephanie Carroll, Nico Christie, Manuel Cortes, Arda Demirci, Melissa Du, Frankie Li, Shuying Luo, Peter Y Wang, Mathew Willows, Feitong Yang, Guangyu Robert Yang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.00114)], 2024.11\n\n- **SocioVerse: A World Model for Social Simulation Powered by LLM Agents and A Pool of 10 Million Real-World Users**\n   *Xinnong Zhang, Jiayu Lin, Xinyi Mou, Shiyue Yang, Xiawei Liu, Libo Sun, Hanjia Lyu, Yihang Yang, Weihong Qi, Yue Chen, Guanying Li, Ling Yan, Yao Hu, Siming Chen, Yu Wang, Xuanjing Huang, Jiebo Luo, Shiping Tang, Libo Wu, Baohua Zhou, Zhongyu Wei.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.10157)], 2025.04\n\n- **Modeling Earth-Scale Human-Like Societies with One Billion Agents**\n   *Haoxiang Guan, Jiyan He, Liyang Fan, Zhenzhen Ren, Shaobin He, Xin Yu, Yuan Chen, Shuxin Zheng, Tie-Yan Liu, Zhen Liu.* [[abs](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2506.12078)], 2025.06\n\n\n\n\n\n\n### 🤖 GUI Agents for Games\n\n- **Lumine: An Open Recipe for Building Generalist Agents in 3D Open Worlds**\n   *Weihao Tan, Xiangyang Li, Yunhao Fang, Heyuan Yao, Shi Yan, Hao Luo, Tenglong Ao, Huihui Li, Hongbin Ren, Bairen Yi, Yujia Qin, Bo An, Libin Liu, Guang Shi.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.08892)], ByteDance Seed 2025.11\n\n- **The Dawn of GUI Agent: A Preliminary Case Study with Claude 3.5 Computer Use**\n   *Siyuan Hu, Mingyu Ouyang, Difei Gao, Mike Zheng Shou.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.10323)], 2024.11\n\n### 🧠 Anthropomorphic Cognition\n\n#### 🎭 Personality Traits\n\n- **Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI**\n   *Sharan Maiya, Henning Bartsch, Nathan Lambert, Evan Hubinger.* [[abs](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2511.01689)], 2025.11\n\n- **PersonaFuse: A Personality Activation-Driven Framework for Enhancing Human-LLM Interactions**\n   *Yixuan Tang, Yi Yang, Ahmed Abbasi.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.07370)]， 2025.09\n\n- **Persona vectors: Monitoring and controlling character traits in language models**\n   *Runjin Chen, Andy Arditi, Henry Sleight, Owain Evans, Jack Lindsey.* [[abs](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Fpersona-vectors)], Anthropic, 2025.08 \n\n-  **Personality Traits in Large Language Models**\n\n   *Mustafa Safdari, Greg Serapio-García, Clément Crepy, Stephen Fitz, Peter Romero, Luning Sun, Marwa Abdulhai, Aleksandra Faust, Maja Matarić.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.00184)], 2023.7\n\n-  **Estimating the Personality of White-Box Language Models.**\n\n   *Saketh Reddy Karra, Son The Nguyen, Theja Tulabandhula.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.12000)], 2022.4\n\n-  **PersonaLLM: Investigating the Ability of GPT-3.5 to Express Personality Traits and Gender Differences.**\n    \n    *Hang Jiang, Xiajie Zhang, Xubo Cao, Jad Kabbara.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.02547)], 2023.5\n\n-  **Does GPT-3 Demonstrate Psychopathy? Evaluating Large Language Models from a Psychological Perspective.**\n\n   *Xingxuan Li, Yutong Li, Shafiq Joty, Linlin Liu, Fei Huang, Lin Qiu, Lidong Bing.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.10529)], 2022.12\n\n-  **Evaluating and Inducing Personality in Pre-trained Language Models.**\n   *Guangyuan Jiang, Manjie Xu, Song-Chun Zhu, Wenjuan Han, Chi Zhang, Yixin Zhu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.07550)], 2022.6\n\n-  **Revisiting the Reliability of Psychological Scales on Large Language Models (Previously: ChatGPT an ENFJ, Bard an ISTJ:Empirical Study on Personalities of Large Language Models.)**\n   *Jen-tse Huang, Wenxuan Wang, Man Ho Lam, Eric John Li, Wenxiang Jiao, Michael R. Lyu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.19926)], 2023.5\n\n-  **Do llms possess a personality? making the mbti test an amazing evaluation for large language models.**\n\n   *Keyu Pan, Yawen Zeng.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.16180)], 2023.7\n\n-  **Can chatgpt assess human personalities? a general evaluation framework.**\n\n   *Haocong Rao, Cyril Leung, Chunyan Miao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.01248)], 2023.3\n\n-  **Who is GPT-3? An Exploration of Personality, Values and Demographics.**\n\n   *Marilù Miotto, Nicola Rossberg, Bennett Kleinberg.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.14338)], 2022.9\n\n-  **Editing Personality for LLMs**\n    \n   *Shengyu Mao, Ningyu Zhang, Xiaohan Wang, Mengru Wang, Yunzhi Yao, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.02168)], 2023.10\n\n\n-  **MBTI Personality Prediction for Fictional Characters Using Movie Scripts**\n    \n    *Yisi Sang, Xiangyang Mou, Mo Yu, Dakuo Wang, Jing Li, Jeffrey Stanton.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.10994)], EMNLP 2022\n\n-  **Unleashing Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration**\n    \n    *Zhenhailong Wang, Shaoguang Mao, Wenshan Wu, Tao Ge, Furu Wei, Heng Ji.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.05300)], 2023.7\n   \n-  **Open Models, Closed Minds? On Agents Capabilities in Mimicking Human Personalities through Open Large Language Models**\n\n   *Lucio La Cava, Davide Costa, Andrea Tagarelli.*, [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.07115)], 2024.01\n\n- **Generative Agent Simulations of 1,000 People**\n   *Joon Sung Park, Carolyn Q. Zou, Aaron Shaw, Benjamin Mako Hill, Carrie Cai, Meredith Ringel Morris, Robb Willer, Percy Liang, Michael S. Bernstein.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.10109)], 2024.11\n\n\n#### 👥 Social Intelligence & Theory-of-Mind\n\n- **SI-Bench: Benchmarking Social Intelligence of Large Language Models in Human-to-Human Conversations**\n   *Shuai Huang, Wenxuan Zhao, Jun Gao.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.23182)], 2025.10\n\n-  **Theory of Mind May Have Spontaneously Emerged in Large Language Models**\n\n   *Michal Kosinski.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2302.02083)], 2023.2\n\n-  **SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents**\n\n   *Xuhui Zhou, Hao Zhu, Leena Mathur, Ruohong Zhang, Haofei Yu, Zhengyang Qi, Louis-Philippe Morency, Yonatan Bisk, Daniel Fried, Graham Neubig, Maarten Sap.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.11667)], 2023.10\n\n- **SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents**\n\n   *Ruiyi Wang, Haofei Yu, Wenxin Zhang, Zhengyang Qi, Maarten Sap, Graham Neubig, Yonatan Bisk, Hao Zhu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.08715)], 2024.03\n\n- **AgentSense: Benchmarking Social Intelligence of Language Agents through Interactive Scenarios**\n   *Xinyi Mou, Jingcong Liang, Jiayu Lin, Xinnong Zhang, Xiawei Liu, Shiyue Yang, Rong Ye, Lei Chen, Haoyu Kuang, Xuanjing Huang, Zhongyu Wei* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.19346)], 2024.10\n\n- **Towards Objectively Benchmarking Social Intelligence for Language Agents at Action Level**\n   *Chenxu Wang, Bin Dai, Huaping Liu, Baoyuan Wang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.05337)], 2024.04\n\n-  **OpenToM: A Comprehensive Benchmark for Evaluating Theory-of-Mind Reasoning Capabilities of Large Language Models.**\n   *Hainiu Xu, Runcong Zhao, Lixing Zhu, Jinhua Du, Yulan He.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.06044)], 2024.02.\n\n-  **Is this the real life? Is this just fantasy? The Misleading Success of Simulating Social Interactions With LLMs**\n   *Xuhui Zhou, Zhe Su, Tiwalayo Eisape, Hyunwoo Kim, Maarten Sap* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.05020)], 2024.03\n\n- **SocialEval: Evaluating Social Intelligence of Large Language Models**\n  *Jinfeng Zhou, Yuxuan Chen, Yihan Shi, Xuanming Zhang, Leqi Lei, Yi Feng, Zexuan Xiong, Miao Yan, Xunzhi Wang, Yaru Cao, Jianing Yin, Shuai Wang, Quanyu Dai, Zhenhua Dong, Hongning Wang, Minlie Huang.* [[abs](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2506.00900)], 2025.06\n\n- **Sotopia-RL: Reward Design for Social Intelligence**\n  *Haofei Yu, Zhengyang Qi, Yining Zhao, Kolby Nottingham, Keyang Xuan, Bodhisattwa Prasad Majumder, Hao Zhu, Paul Pu Liang, Jiaxuan You.* [[abs](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2508.03905)], 2025.08\n\n#### 🧠 Others \u003C!---others-1-->\n\n-  **Role-Play with Large Language Models**\n   *Murray Shanahan, Kyle McDonell, and Laria Reynolds.*[[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.16367)], 2023.5\n\n-  **Consciousness in artificial intelligence: Insights from the science of consciousness.**\n   *Patrick Butlin, Robert Long, Eric Elmoznino, Yoshua Bengio, Jonathan Birch, Axel Constant, George Deane, Stephen M. Fleming, Chris Frith, Xu Ji, Ryota Kanai, Colin Klein, Grace Lindsay, Matthias Michel, Liad Mudrik, Megan A. K. Peters, Eric Schwitzgebel, Jonathan Simon, Rufin VanRullen.*[[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.08708)], 2023.8\n\n-  **The Self-Perception and Political Biases of ChatGPT.**\n\n   *Jérôme Rutinowski, Sven Franke, Jan Endendyk, Ina Dormuth, Markus Pauly.*, 2023.4. Tags: personality traits, political biases, dark personality traits.\n\n-  **The political ideology of conversational AI: Converging evidence on ChatGPT's pro-environmental, left-libertarian orientation.**\n\n   *Jochen Hartmann, Jasper Schwenzow, Maximilian Witte.*, 2023.1. Tags: political biases.\n\n-  **Inducing anxiety in large language models increases exploration and bias.**\n\n   *Julian Coda-Forno, Kristin Witte, Akshay K. Jagadish, Marcel Binz, Zeynep Akata, Eric Schulz.*, 2023.4. Tags: anxiety.\n\n-  **Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology View.**\n\n   *Jintian Zhang, Xin Xu, Shumin Deng.*, 2023.10. Tags: social psychology.\n\n- **REALTALK: A 21-Day Real-World Dataset for Long-Term Conversation**\n   *Dong-Ho Lee, Adyasha Maharana, Jay Pujara, Xiang Ren, Francesco Barbieri.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13270)], 2025.2\n\n### 🌐 Others \u003C!---others-2-->\n\n#### 📂 Resources\n\n- **Model: CoSER-8B, CoSER-70B**\n   *Xintao Wang, et.al.* [[huggingface](https:\u002F\u002Fgithub.com\u002FNeph0s\u002FCOSER)]\n\n- **Model: Deepsex**\n   *ValueFX9507.* [[huggingface](https:\u002F\u002Fhuggingface.co\u002FValueFX9507\u002FTifa-Deepsex-14b-CoT-GGUF-Q4)]\n\n- **Benchmark: RPBench Leaderboard**\n   *Boson AI.* [[leaderboard](https:\u002F\u002Fboson.ai\u002Frpbench\u002F)]\n\n- **Benchmark: Fiction.liveBench**\n   *Fiction.live.* [[benchmark](https:\u002F\u002Ffiction.live\u002Fstories\u002FFiction-liveBench-Mar-14-2025\u002FoQdzQvKHw8JyXbN87\u002Fhome)]\n\n- **Model: Higgs-Llama-3-70B**\n   *Boson AI.* [[huggingface](https:\u002F\u002Fhuggingface.co\u002Fbosonai\u002FHiggs-Llama-3-70B)]\n\n#### 📰 Blogs\n\n- **[[Emotion concepts and their function in a large language model](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Femotion-concepts-function)]**, Anthropic Blog, 2026.04\n\n- **[[The assistant axis: situating and stabilizing the character of large language models](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Fassistant-axis)]**, Anthropic Blog, 2026.01\n\n- **[[Persona vectors: Monitoring and controlling character traits in language models](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Fpersona-vectors)]**, Anthropic Blog, 2025.08\n- \n- **[[Character.ai 赛道的真相 (The Truth About the Character.ai Track)](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002FH2KNDGRNHktHiQc3sayFsA)]**, 2024.08\n\n- **[[李沐：创业一年，人间三年！(Mu Li (Boson AI): One Year of Entrepreneurship, Three Years in the Human World!)](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002F2lbCMo64-nU5yRz1cLQxYA)]**, 2024.08\n\n- **[[盘点ACL 2024 角色扮演方向都发了啥 (A Review of Publications in Role-Playing Direction at ACL 2024)](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002FE5qp5YPYPVaLM07OumDTRw)]**, 2024.07\n\n- **[[角色扮演大模型的碎碎念 (Ramblings on Large Language Models for Role-Playing)](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002FyoM-srJYGGfyd1VXirg_Hg)]**, 2024.03\n\n- **[[聊一聊做角色扮演大模型的经验 (Discussing Experiences in Developing Role-Playing Large Language Models)](https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F3641608640)]**, 2024.10\n\n- **[[角色扮演大模型技术分享 (Technical Insights on Role-Playing Large Language Models)](https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F685823865)]**, 2024.03\n\n- **[[角色扮演大模型技术分享2-超拟人模型的困境 (Technical Insights on Role-Playing Large Language Models 2 - The Dilemma of Super-Anthropomorphic Models)](https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F719135803)]**, 2024.09\n\n- **[[角色扮演大模型技术分享3-拟人能力提升&数据合成pipeline (Technical Insights on Role-Playing Large Language Models 3 - Improving Anthropomorphic Capabilities & Data Synthesis Pipeline)](https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F719772276)]**, 2024.09\n\n\n#### 📄 Papers\n\n- **One Thousand and One Pairs: A \"novel\" challenge for long-context language models**\n   *Marzena Karpinska, Katherine Thai, Kyle Lo, Tanya Goyal, Mohit Iyyer.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.16264)], 2024.06\n\n- **NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens**\n   *Cunxiang Wang, Ruoxi Ning, Boqi Pan, Tonghui Wu, Qipeng Guo, Cheng Deng, Guangsheng Bao, Xiangkun Hu, Zheng Zhang, Qian Wang, Yue Zhang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.12766)], 2024.03\n\n\n- **Visual-RolePlay: Universal Jailbreak Attack on MultiModal Large Language Models via Role-playing Image Character**\n   *Siyuan Ma, Weidi Luo, Yu Wang, Xiaogeng Liu.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.20773)], 2024.06\n\n- **AI Cat Narrator: Designing an AI Tool for Exploring the Shared World and Social Connection with a Cat**\n   *Zhenchi Lai, Janet Yi-Ching Huang, Rung-Huei Liang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.06192)], 2024.06\n\n- **THEANINE: Revisiting Memory Management in Long-term Conversations with Timeline-augmented Response Generation**\n  *Seo Hyun Kim, Kai Tzu-iunn Ong, Taeyoon Kwon, Namyoung Kim, Keummin Ka, SeongHyeon Bae, Yohan Jo, Seung-won Hwang, Dongha Lee, Jinyoung Yeo.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.10996)], 2024.06\n\n- **ESC-Eval: Evaluating Emotion Support Conversations in Large Language Models**\n   *Haiquan Zhao, Lingyu Li, Shisong Chen, Shuqi Kong, Jiaan Wang, Kexin Huang, Tianle Gu, Yixu Wang, Dandan Liang, Zhixu Li, Yan Teng, Yanghua Xiao, Yingchun Wang.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.14952)], 2024.06\n\n- **PATIENT-Ψ: Using Large Language Models to Simulate Patients for Training Mental Health Professionals**\n   *Ruiyi Wang, Stephanie Milani, Jamie C. Chiu, Jiayin Zhi, Shaun M. Eack, Travis Labrum, Samuel M. Murphy, Nev Jones, Kate Hardy, Hong Shen, Fei Fang, Zhiyu Zoey Chen.* [[abs](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.19660)], 2024.05\n\n- **CharacterMeet: Supporting Creative Writers' Entire Story Character Construction Processes Through Conversation with LLM-Powered Chatbot Avatars**\n   *Hua Xuan Qin, Shan Jin, Ze Gao, Mingming Fan, Pan Hui.' [[abs](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3613904.3642105)], CHI 2024.\n\n- **Stephanie: Step-by-Step Dialogues for Mimicking Human Interactions in Social Conversations**\n   *Hao Yang, Hongyuan Lu, Xinhua Zeng, Yang Liu, Xiang Zhang, Haoran Yang, Yumeng Zhang, Yiran Wei, Wai Lam.* [[abs](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2407.04093)], 2024.07\n\n\n---\n\n## 💡  Contribution\n\n### Contributing to this repo\n\n🤲\" **Join us in improving this repository!** Spotted any notable works we might have missed? We welcome your additions. Every contribution counts!   \"\n\n### Contributors\n\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FNeph0s\u002Fawesome-llm-role-playing-with-persona\u002Fgraphs\u002Fcontributors\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FNeph0s_awesome-llm-role-playing-with-persona_readme_6d8b2ac02506.png\" \u002F>\n\u003C\u002Fa>\n","# 关于角色扮演语言模型的论文与资源\n\n[![Awesome](https:\u002F\u002Fcdn.rawgit.com\u002Fsindresorhus\u002Fawesome\u002Fd7305f38d29fed78fa85652e3a63e154dd8e8829\u002Fmedia\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FNeph0s\u002Fawesome-llm-role-playing-with-persona) \n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-yellow.svg)](https:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT) \n![最近提交](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flast-commit\u002FNeph0s\u002Fawesome-llm-role-playing-with-persona.svg)\n\n---\n\n本仓库主要聚焦于**角色扮演**，例如虚构角色、名人及历史人物等。尽管角色扮演语言代理与多智能体系统、长上下文模型等诸多研究领域相关，但我们无法保证这些领域的论文都会被收录。\n\n*以下是一些您可能感兴趣的其他论文列表：*\n\n🤖 ![最近提交](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flast-commit\u002Fzjunlp\u002FLLMAgentPapers.svg)**[LLM 代理论文](https:\u002F\u002Fgithub.com\u002Fzjunlp\u002FLLMAgentPapers)：** 大型语言模型代理领域的必读论文。\n\n## 🔔 最新消息\n \n- **2024-10-09 🎉 我们的综述论文\"*从人格到个性化：关于角色扮演语言代理的综述\"* [[arxiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.18231)] 已被 TMLR 接受！**\n\n- **2024-06-28 我们重新整理了此仓库，更加专注于角色扮演代理。**\n\n- **2024-04-30 我们非常高兴地宣布我们的综述论文：\"*从人格到个性化：关于角色扮演语言代理的综述\"* [[arxiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.18231)]**\n\n- **2024-04-17 我们正在寻找对该主题研究感兴趣的合作者。请通过电子邮件联系我们：xtwang21@m.fudan.edu.cn**\n\n- **2023-10-30 我们创建了这个仓库，用于维护关于*具有指定人格的大语言模型角色扮演*的论文列表。**\n\n## 📜 内容\n\n- [LLM 角色扮演\u002F人格论文与资源](#llm-role-playingpersona-papers-and-resources)\n  - [🔔 最新消息](#-news)\n  - [📜 内容](#content)\n    - [🧙 角色扮演](#-role-playing)\n      - [🧙 角色](#-characters)\n      - [👨‍👩‍👧‍👦 人口统计学](#️-demographics)\n      - [🤖 个性化](#-personalization)\n    - [👥 多智能体](#-multi-agents)\n    - [🤖 用于游戏的 GUI 代理](#-gui-agents-for-games)\n    - [🧠 拟人化认知](#-anthropomorphic-cognition)\n      - [🎭 性格特质](#-personality-traits)\n      - [👥 社会智能与心智理论](#-social-intelligence--theory-of-mind) \n      - [🧠 其他](#-others-1)\n    - [🌐 其他](#-others-2)\n      - [📂 资源](#-resources)\n      - [📰 博客](#-blogs)\n      - [📄 论文](#-papers)\n  - [💡 贡献](#--contribution)\n    - [为本仓库做贡献](#contributing-to-this-repo)\n    - [贡献者](#contributors)\n\n\n\n### 🧙 角色扮演\n\n#### 🧙 角色\n\n*角色扮演 LLM 快速简介*\n\n-  **从人格到个性化：关于角色扮演语言代理的综述。**\n   \n   *Jiangjie Chen, Xintao Wang, Rui Xu, Siyu Yuan, Yikai Zhang, Wei Shi, Jian Xie, Shuang Li, Ruihan Yang, Tinghui Zhu, Aili Chen, Nianqi Li, Lida Chen, Caiyu Hu, Siye Wu, Scott Ren, Ziquan Fu, Yanghua Xiao.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.18231)], TMLR 2024年4月\n\n-  **CoSER：协调基于 LLM 的既定角色的人格模拟**\n\n   *Xintao Wang, Heng Wang, Yifei Zhang, Xinfeng Yuan, Rui Xu, Jen-tse Huang, Siyu Yuan, Haoran Guo, Jiangjie Chen, Wei Wang, Yanghua Xiao, Shuchang Zhou.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09082)], ICML 2025\n\n-  **RoleLLM：大型语言模型角色扮演能力的基准测试、激发与提升。**\n\n   *Zekun Moore Wang, Zhongyuan Peng, Haoran Que, Jiaheng Liu, Wangchunshu Zhou, Yuhan Wu, Hongcheng Guo, Ruitong Gan, Zehao Ni, Man Zhang, Zhaoxiang Zhang, Wanli Ouyang, Ke Xu, Wenhu Chen, Jie Fu, Junran Peng.*, [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.00746)], ACL 2024发现\n\n-  **ChatHaruhi：通过大型语言模型在现实中复活动漫角色。**\n\n   *Cheng Li, Ziang Leng, Chenxi Yan, Junyi Shen, Hao Wang, Weishi MI, Yaying Fei, Xiaoyang Feng, Song Yan, HaoSheng Wang, Linkang Zhan, Yaokai Jia, Pingyu Wu, Haozhen Sun.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.09597)], 2023年8月\n\n-  **Character-LLM：一种可训练的角色扮演代理。**\n\n   *Yunfan Shao, Linyang Li, Junqi Dai, Xipeng Qiu.*, [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.10158)], EMNLP 2023\n\n-  **CharacterEval：中文角色扮演对话代理评估基准。**\n   *Quan Tu, Shilong Fan, Zihang Tian, Rui Yan.*, [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.01275)], ACL 2024\n\n-  **InCharacter：通过心理访谈评估角色扮演代理中的人格一致性。**\n\n   *Xintao Wang, Yunze Xiao, Jen-tse Huang, Siyu Yuan, Rui Xu, Haoran Guo, Quan Tu, Yaying Fei, Ziang Leng, Wei Wang, Jiangjie Chen, Cheng Li, Yanghua Xiao.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.17976)], ACL 2024\n\n- **BookWorld：从小说到互动代理社会，用于创意故事生成**\n\n   *Yiting Ran, Xintao Wang, Tian Qiu, Jiaqing Liang, Yanghua Xiao, Deqing Yang.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.14538)], ACL 2025\n\n- **性格即命运：大型语言模型能否在角色扮演中模拟由人格驱动的决策？**\n\n   *Rui Xu, Xintao Wang, Jiangjie Chen, Siyu Yuan, Xinfeng Yuan, Jiaqing Liang, Zulong Chen, Xiaoqing Dong, Yanghua Xiao.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.12138)], EMNLP 2025发现\n\n- **CharacterBench：大型语言模型角色自定义能力的基准测试**\n   \n   *Jinfeng Zhou, Yongkang Huang, Bosi Wen, Guanqun Bi, Yuxuan Chen, Pei Ke, Zhuang Chen, Xiyao Xiao, Libiao Peng, Kuntian Tang, Rongsheng Zhang, Le Zhang, Tangjie Lv, Zhipeng Hu, Hongning Wang, Minlie Huang* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.11912)], AAAI 2025\n\n-  **OmniCharacter：迈向沉浸式角色扮演代理，实现无缝的语音-语言人格交互**\n\n   *Haonan Zhang, Run Luo, Xiong Liu, Yuchuan Wu, Ting-En Lin, Pengpeng Zeng, Qiang Qu, Feiteng Fang, Min Yang, Lianli Gao, Jingkuan Song, Fei Huang, Yongbin Li.* [[摘要](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2505.20277)], ACL 2025\n\n- **人格向量：监控和控制语言模型中的角色特质**\n   *Runjin Chen, Andy Arditi, Henry Sleight, Owain Evans, Jack Lindsey.* [[摘要](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Fpersona-vectors)], Anthropic 博客，2025年8月\n\n\n- **CPO：通过比较策略优化解决角色扮演对话中的奖励模糊性**\n   *Xinge Ye, Rui Wang, Yuchuan Wu, Victor Ma, Feiteng Fang, Fei Huang, Yongbin Li.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.09074)], Tongyi Lab，2025年8月\n\n- **CogDual：通过基于隐式规则的强化学习增强 LLM 的双重认知**\n   *Cheng Liu, Yifei Lu, Fanghua Ye, Jian Li, Xingyu Chen, Feiliang Ren, Zhaopeng Tu, Xiaolong Li.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.17147)], 2025年7月\n\n- **助手轴：定位并稳定语言模型的默认人格**\n   *Christina Lu, Jack Gallagher, Jonathan Michala, Kyle Fish, Jack Lindsey.* [[摘要](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2601.10387)], 2026年1月\n\n- **情感概念及其在大型语言模型中的作用**\n   *尼古拉斯·索弗罗尼耶、艾萨克·考瓦尔、威廉·桑德斯、陈润锦、汤姆·赫尼根、萨莎·海德里、克雷格·西特罗、亚当·皮尔斯、朱利叶斯·唐、韦斯·格尼、乔舒亚·巴特森、萨姆·齐默曼、凯莉·里瓦尔、凯尔·费什、克里斯·奥拉、杰克·林赛。* [[网页](https:\u002F\u002Ftransformer-circuits.pub\u002F2026\u002Femotions\u002Findex.html)]，2026年4月\n\n\n*全部*\n*已确立角色*\n\n- **HER：面向LLM角色扮演的人类化推理与强化学习**\n   *杜成宇、王新涛、陈爱丽、李伟远、徐锐、刘俊腾、黄子山、田荣、孙子俊、李宇豪、冯立恒、丁德明、赵鹏宇、肖阳华。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.21459)]，ACL 2026会议成果\n\n- **HumanLLM：基于人类认知模式的LLM拟人化评测与改进**\n   *王新涛、杨健、李伟远、谢睿、黄仁哲、高军、黄帅、康月萍、苟立元、冯宏伟、肖阳华。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.10198)]，ACL 2026\n\n\n- **MOA：面向角色扮演智能体的多目标对齐**\n   *廖崇华、王科、吴宇川、黄飞、李永斌。* [[摘要](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2512.09756)]，2025年12月\n\n- **好得不像坏：论LLM在反派角色扮演中的失败**\n   *易子浩、蒋庆轩、马若天、陈星宇、曲扬、王梦茹、叶芳华、沈颖、涂兆鹏、李晓龙、林纳斯。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.04962)]，2025年11月\n\n- **CPO：通过比较策略优化解决角色扮演对话中的奖励歧义问题**\n   *叶新歌、王睿、吴宇川、马维克、方飞腾、黄飞、李永斌。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.09074)]，通义实验室，2025年8月\n\n- **RMTBench：通过多轮以用户为中心的角色扮演评测LLM**\n   *向浩、唐天一、苏洋、于博文、杨安、黄飞、张一昌、陆耀杰、林洪宇、韩先培、周景仁、林俊阳、孙乐。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.20352)]，2025年7月\n\n- **CogDual：利用基于隐式规则的奖励强化学习提升LLM的双重认知能力**\n   *刘程、陆一飞、叶芳华、李建、陈星宇、任飞亮、涂兆鹏、李晓龙。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.17147)]，2025年7月\n\n- **识别出戏行为：开放生成中人格一致性的原子级评估**\n   *申智秀、吴珠贤、金恩秀、宋浩允、欧艾丽丝。* [[摘要](https:\u002F\u002Faclanthology.org\u002F2025.findings-acl.1349\u002F)]，ACL 2025会议成果\n\n- **Crab：一款具有评估基准的新颖可配置角色扮演LLM**\n   *何凯、黄宇诚、王文清、冉德隆、盛东明、黄俊轩、林奇卡、许嘉兴、刘文强、冯梦玲。* [[摘要](https:\u002F\u002Faclanthology.org\u002F2025.acl-long.731\u002F)]，ACL 2025\n\n- **测试时匹配：解耦基于LLM的角色扮演语言智能体中的个性、记忆与语言风格**\n\n   *詹小宇、傅欣宇、孙浩、李元琪、郭杰、郭燕文。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.16799)]，2025年7月\n\n- **以角色思维：用角色意识推理推进角色扮演智能体的发展**\n\n   *唐义宏、陈可海、杨木云、牛正宇、李静、赵铁军、张敏。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.01748)]，2025年6月\n\n\n- **BookWorld：从小说到互动式智能体社会，用于创意故事生成**\n\n   *冉怡婷、王新涛、邱天、梁佳青、肖阳华、杨德庆。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.14538)]，ACL 2025\n\n- **OmniCharacter：迈向沉浸式角色扮演智能体——实现语音与语言人格的无缝交互**\n\n   *张浩楠、罗润、刘雄、吴宇川、林廷恩、曾鹏鹏、屈强、方飞腾、杨敏、高莲莉、宋京宽、黄飞、李永斌。* [[摘要](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2505.20277)]，ACL 2025\n\n- **PsyMem：面向高级角色扮演LLM的细粒度心理对齐与显式记忆控制**\n\n   *程锡龙、秦云霄、谭雨婷、李正南、王晔、肖洪江、张源。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.12814)]，2025年5月\n\n- **RAIDEN-R1：通过可验证奖励的GRPO提升LLM的角色意识**\n   *王宗胜、孙凯丽、吴博文、于群、李英、王宝勋。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.10218)]，2025年5月\n\n- **ChARM：面向高级角色扮演语言智能体的基于角色的行为自适应奖励建模**\n   *方飞腾、林廷恩、吴宇川、刘雄、黄翔、陈定威、叶静、张浩楠、朱亮、哈米德·阿利内贾德-罗克尼、杨敏、黄飞、李永斌。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.23923)]，2025年5月\n\n- **角色逻辑的编码化在角色扮演中的应用**\n   *彭乐天、尚景波。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.07705)]，2025年5月\n\n- **当哈利遇到超人：对话伙伴在基于人格的对话生成中的作用**\n   *丹妮拉·奥奇平蒂、马尔科·格里尼、马尔维娜·尼西姆。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.24613)]，2025年5月\n\n- **CoSER：协调基于LLM的既定角色人格模拟**\n\n   *王新涛、王恒、张一飞、袁信峰、徐睿、黄仁哲、袁思宇、郭浩然、陈江杰、王伟、肖阳华、周书畅。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09082)]，ICML 2025\n\n- **迈向RPA评估的设计指南：基于大型语言模型的角色扮演智能体综述**\n   *陈超然、姚炳生、邹瑞石、华文悦、吕伟民、叶艳芳、李家俊、王大阔。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13012)]，2025年2月\n\n- **RPGBENCH：将大型语言模型作为角色扮演游戏引擎进行评估**\n   *余鹏飞、沈东明、孟思琳、李在源、尹伟素、崔瑶云、徐振林、朱毅、施兴健、穆力、亚历克斯·斯莫拉。* [[摘要](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2502.00595)]，2025年2月\n\n- **推理并不一定提升角色扮演能力**\n   *冯夏冲、窦隆旭、孔凌鹏。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.16940)]，2025年2月\n\n- **CharacterBox：在文本虚拟世界中评估LLM的角色扮演能力**\n   *王磊、连建勋、黄毅、戴彦琦、李浩轩、陈旭、谢兴、温继荣。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.05631)]，2024年12月\n\n- **揭示并缓解LLM角色扮演中检测角色知识错误的挑战**\n   *张文远、聂帅义、盛嘉伟、张泽峰、张兴华、何勇泉、刘廷文。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11726)]，2024年9月\n\n- **减轻虚构角色扮演中的幻觉现象**\n\n   *纳菲斯·萨德克、谢舟航、姜炳奎、普拉里特·兰巴、高翔、朱利安·麦考利。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.17260)] [[数据集](https:\u002F\u002Fgithub.com\u002FNafisSadeq\u002Frolefact)]，EMNLP 2024会议成果\n\n- **量化与优化基于人格的角色扮演中的全局忠实度**\n\n   *彭乐天、尚景波。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.07726)] [[数据集](https:\u002F\u002Fgithub.com\u002FKomeijiForce\u002FActive_Passive_Constraint_Koishiday_2024)]，NeurIPS 2024\n\n-  **CharacterGLM：利用大型語言模型定制中文對話式AI角色。**\n   *周金峰、陳莊、萬大震、文博思、宋毅、余繼帆、黃永康、彭利標、楊嘉明、肖西堯、薩漢德·薩布爾、張曉涵、侯文靜、張一佳、董宇霄、唐傑、黃民烈*，[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.16832)]，EMNLP 2024，產業專題\n\n-  **PIPPA：一個部分合成的對話數據集**\n   *蒂爾·戈斯林、阿爾平·戴爾、鄭銀河* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.05884)]，2023年8月\n\n- **性格即命運：大型語言模型能否在角色扮演中模擬由人格特質驅動的決策？**\n\n   *許睿、王新濤、陳江杰、袁思宇、袁信豐、梁家慶、陳祖龍、董曉青、肖陽華* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.12138)]，EMNLP 2025研究成果\n\n\n-  **大型語言模型是所有角色的疊加：通過自我對齊實現任意角色扮演**\n   *陸克明、于博文、周暢、周敬仁* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.12474)]，2024年1月\n\n-  **大型語言模型與哈利·波特相遇：用於使對話代理與角色對齊的雙語數據集。**\n   *陳諾、王彥、蔣海雲、蔡登、李雨涵、陳子陽、王隆悅、李佳* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.06869)]，2022年11月\n\n- **通過小說作品中的角色剖析評估大型語言模型的角色理解能力**\n   *袁信豐、袁思宇、崔雨涵、林天賀、王新濤、許睿、陳江杰、楊德清* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.12726)]，2024年4月\n\n\n- **從角色扮演到戲劇互動：一種基於LLM的解決方案**\n   *吳偉奇、吳紅秋、江來、劉星源、洪佳樂、趙海、張敏* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.14231)]，ACL 2024研究成果\n\n- **TimeChara：評估角色扮演型大型語言模型的時點角色幻覺能力**\n   *安在宇、李泰賢、林俊英、金振和、尹尚斗、李花蘭、金根熙* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.18027)]，ACL 2024研究成果\n\n\n-  **“讓你的角色講述他們的故事”：一個以角色為中心的敘事理解數據集**\n   *法澤·布拉曼、黃萌、奧伊溫德·塔夫約德、趙超、姆里納亞·薩昌、斯尼格達·查圖維迪* [[摘要](https:\u002F\u002Faclanthology.org\u002F2021.findings-emnlp.150\u002F)]，EMNLP 2021研究成果\n\n\n-  **CharacterEval：一個用於評估角色扮演對話代理的中文基準測試**\n   *涂泉、范士龍、田子航、嚴瑞* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.01275)]，ACL 2024\n\n\n-  **Neeko：利用動態LoRA高效實現多角色扮演代理**\n   *余小燕、羅通旭、魏一凡、雷方宇、黃一鳴、彭浩、朱立煌* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.13717)]，2024年2月\n\n-  **RoleEval：一個用於大型語言模型的角色評價雙語基準測試**\n   *沈天昊、李孫、涂泉、熊德義* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.16132)]，2023年12月\n\n-  **ERABAL：通過邊界感知學習增強角色扮演代理**\n   *唐義宏、歐嬌、劉徹、張福正、張迪、蓋坤* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.14710)]，2024年9月\n\n- **SocialBench：角色扮演對話代理的社交性評估**\n   *陳宏展、陳和宏、閻明、徐文深、高興、沈偉洲、全曉軍、李晨亮、張濟、黃飛、周景仁* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.13679)]，2024年3月\n\n- **捕捉心靈，而非僅僅文字：利用人格指標數據增強角色扮演語言模型**\n   *冉怡婷、王新濤、許睿、袁信豐、梁家慶、肖陽華、楊德清* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.18921)]，EMNLP 2024研究成果\n\n- **MMRole：開發與評估多模態角色扮演代理的綜合框架**\n   *戴艷琪、胡歡然、王磊、金聖傑、陳旭、陸志武* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.04203)]，2024年8月\n\n- **揭示檢測LLM角色扮演中角色知識錯誤的挑戰**\n   *張文遠、盛嘉偉、聶帥毅、張澤豐、張興華、何永泉、劉廷文* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11726)]，2024年9月\n\n- **超越對話：面向通用角色扮演語言模型的輪廓-對話對齊框架**\n   *俞葉勇、俞潤生、魏浩傑、張贊秋、錢權* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.10903)]，2024年8月\n\n- **PRODIGy：基於個人資料的對話生成數據集**\n  *丹妮拉·奧奇平蒂、塞拉·希內姆·泰基羅格魯、馬可·圭里尼* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.05195)] [[數據集](https:\u002F\u002Fgithub.com\u002FLanD-FBK\u002Fprodigy-dataset)]，NAACL 2024研究成果\n\n- **CharacterBench：大型語言模型角色自定義的基準測試**\n   *周金峰、黃永康、文博思、畢冠群、陳宇軒、柯沛、陳莊、肖西堯、彭利標、唐坤田、張榮生、張樂、呂唐傑、胡志鵬、王洪寧、黃民烈* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.11912)]，AAAI 2025\n\n*合成角色*\n\n- **LLM生成的人格特質是一種有陷阱的承諾**\n   *李昂、陳浩哲、南孔洪弘、彭天儀* [[摘要](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2503.16527)]，2025年3月\n\n- **RoleMRC：一個細粒度的複合基準測試，用於角色扮演和指令遵循**\n   *盧俊如、李家政、沈國東、桂琳、安思宇、何玉蘭、殷迪、孫星* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.16527)]，2025年2月\n\n\n- **OpenCharacter：使用大規模合成人格訓練可定制的角色扮演LLM**\n   *王曉陽、張宏明、葛濤、于文浩、于典、于東* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.15427)]，2025年1月\n\n- **以10億個人格擴展合成數據的創建**\n   *陳鑫、王曉陽、于典、米海濤、于東* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.20094)]，2024年6月\n\n- **Stark：結合人格常識知識的社交長時多模態對話**\n   *李永俊、李度鍾、尹俊英、吳京珍、高炳洙、玄鐘煥、崔浩鎮* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.03958)]，EMNLP 2024\n\n- **PersonaGym：評估人格代理與LLM**\n   *維奈·塞繆爾、亨利·彭佐、周悅、謝拉斯·喬杜里、阿什溫·卡利揚、坦邁·拉吉普羅希特、阿米特·德什潘德、卡爾蒂克·納拉西曼、維什瓦克·穆拉哈里* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.18416)]，2024年10月\n\n#### 👨‍👩‍👧‍👦 人口統計學\n\n- **通過人格選擇及價值—信念—規範推理使大型語言模型與人類意見對齊**\n   *杜宣龍、川口健二、簡敏妍、陳南茜* [[摘要](https:\u002F\u002Faclanthology.org\u002F2025.coling-main.172.pdf)]，COLING 2025\n\n-  **專家提示：指導大型語言模型成為卓越的專家**\n\n   *徐本峰、楊安、林俊洋、王泉、周昌、張永東和毛振東* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.14688)]，2023年5月\n\n-  **ChatGPT中的毒性：分析賦予人格的語言模型**\n\n*Ameet Deshpande、Vishvak Murahari、Tanmay Rajpurohit、Ashwin Kalyan、Karthik Narasimhan.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.05335)]，2023年4月\n\n-  **通过角色扮演提示提升零样本推理能力**\n   *孔奥博、赵世万、陈浩、李奇成、秦勇、孙瑞琪、周欣。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.07702)]，2023年8月\n\n-  **偏见根深蒂固：赋予人格的大语言模型中的隐性推理偏见**\n   *Shashank Gupta、Vaishnavi Shrivastava、Ameet Deshpande、Ashwin Kalyan、Peter Clark、Ashish Sabharwal、Tushar Khot* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.04892)]，2023年11月\n\n-  **“乐于助人的助手”是大型语言模型的最佳角色吗？系统性评估系统提示中的社会角色**\n   *郑明谦、裴佳欣、David Jurgens* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.10054)]，2023年11月\n\n-  **CoMPosT：刻画与评估大语言模型模拟中的漫画式表现**\n   *Myra Cheng、Tiziano Piccardi、Diyi Yang。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.11501)]，2023年10月\n\n-  **解密预训练语言模型中的刻板印象**\n   *Weicheng Ma、Henry Scheible、Brian Wang、Goutham Veeramachaneni、Pratim Chowdhary、Alan Sun、Andrew Koulogeorge、Lili Wang、Diyi Yang、Soroush Vosoughi。* [[摘要](https:\u002F\u002Faclanthology.org\u002F2023.emnlp-main.697\u002F)]，EMNLP Findings 2023\n\n-  **CultureLLM：将文化差异融入大型语言模型**\n   *Cheng Li、Mengzhou Chen、Jindong Wang、Sunayana Sitaram、Xing Xie。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.10946)]，2024年2月\n\n- **通过角色选择及价值–信念–规范推理，使大型语言模型与人类观点对齐**\n   *Xuan Long Do、Kenji Kawaguchi、Min-Yen Kan、Nancy Chen。* [[摘要](https:\u002F\u002Faclanthology.org\u002F2025.coling-main.172.pdf)]，COling 2025\n\n#### 🤖 个性化\n\n- **了解我，回应我：大规模动态用户画像与个性化响应的大型语言模型基准测试**\n   *Bowen Jiang、Zhuoqun Hao、Young-Min Cho、Bryan Li、Yuan Yuan、Sihao Chen、Lyle Ungar、Camillo J. Taylor、Dan Roth* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.14225)]，2025年4月\n\n- **更深入地洞察您的用户：面向动态人格建模的定向人格精炼**\n   *Aili Chen、Chengyu Du、Jiangjie Chen、Jinghan Xu、Yikai Zhang、Siyu Yuan、Zulong Chen、Liangyue Li、Yanghua Xiao* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.11078)]，2025年2月\n\n- **大型语言模型离成为我们的数字孪生还有多远？基于人格的行为链仿真基准测试**\n   *Rui Li、Heming Xia、Xinfeng Yuan、Qingxiu Dong、Lei Sha、Wenjie Li、Zhifang Sui* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.14642)]，2025年2月\n\n-  **LaMP：当大型语言模型遇上个性化**\n   *Alireza Salemi、Sheshera Mysore、Michael Bendersky、Hamed Zamani* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.11406)]，2023年4月。\n\n- **大型语言模型能多好地模仿我们？用ECHO评估AI聊天机器人的角色扮演能力**\n   *Man Tik Ng、Hui Tung Tse、Jen-tse Huang、Jingjing Li、Wenxuan Wang、Michael R. Lyu。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.13957)]，2024年4月\n\n-  **关于大型语言模型向数据驱动型人格的可控性**\n   *Junyi Li、Ninareh Mehrabi、Charith Peris、Palash Goyal、Kai-Wei Chang、Aram Galstyan、Richard Zemel、Rahul Gupta* [[摘要](https:\u002F\u002Faclanthology.org\u002F2024.naacl-long.405\u002F)]，NAACL 2024。\n\n- **展示而非告知：以实证反馈对齐语言模型**\n   *Omar Shaikh、Michelle Lam、Joey Hejna、Yijia Shao、Michael Bernstein、Diyi Yang* [[摘要](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2406.00888)]，2024年6月\n\n- **PersLLM：一种面向大型语言模型的人格化训练方法**\n   *Zheni Zeng、Jiayi Chen、Huimin Chen、Yukun Yan、Yuxuan Chen、Zhenghao Liu、Zhiyuan Liu、Maosong Sun* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.12393)]，2024年7月\n\n-  **LiveChat：从直播中自动构建的大规模个性化对话数据集**\n\n   *Jingsheng Gao、Yixin Lian、Ziyi Zhou、Yuzhuo Fu、Baoyuan Wang。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.08401)]，2023年6月。标签：主播人格。\n\n-  **COSPLAY：基于概念集的双方面人格引导式个性化对话生成**\n   *Jingsheng Gao、Yixin Lian、Ziyi Zhou、Yuzhuo Fu、Baoyuan Wang。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.00872)]，SIGIR 2022。\n\n-  **个性化对话代理：我养了一只狗，您也有宠物吗？** \n   *Saizheng Zhang、Emily Dinan、Jack Urbanek、Arthur Szlam、Douwe Kiela、Jason Weston。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.07243)]，ACL 2018。\n\n-  **MPCHAT：迈向多模态人格基底对话**\n   *Jaewoo Ahn、Yeda Song、Sangdoo Yun、Gunhee Kim。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.17388)]，ACL 2023。标签：多模态人格。\n\n\n-  **基于交互学习（IGL）的个性化奖励学习**\n   *Jessica Maghakian、Paul Mineiro、Kishan Panaganti、Mark Rucker、Akanksha Saran、Cheng Tan* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.15823)]，2022年11月\n\n-  **当大型语言模型遇见个性化：挑战与机遇的视角**\n   *Jin Chen、Zheng Liu、Xu Huang、Chenwang Wu、Qi Liu、Gangwei Jiang、Yuanhao Pu、Yuxuan Lei、Xiaolong Chen、Xingmei Wang、Defu Lian、Enhong Chen* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.16376)]，2023年7月，综述\n\n-  **超越聊天机器人：ExploreLLM用于结构化思考与个性化模型响应**\n   *Xiao Ma、Swaroop Mishra、Ariel Liu、Sophie Su、Jilin Chen、Chinmay Kulkarni、Heng-Tze Cheng、Quoc Le、Ed Chi* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.00763)]，2023年12月\n\n- **PersonalityChat：结合事实与特质的个性化对话建模的对话蒸馏**\n   *Ehsan Lotfi、Maxime De Bruyn、Jeska Buhmann、Walter Daelemans。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.07363)]，2024年1月\n\n- **超越人口统计学：利用人类信念网络对齐角色扮演型大语言模型代理**\n   *Yun-Shiuan Chuang、Krirk Nirunwiroj、Zach Studdiford、Agam Goyal、Vincent V. Frigo、Sijia Yang、Dhavan Shah、Junjie Hu、Timothy T. Rogers。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.17232)]，2024年6月\n\n- **1,000人的生成式代理人模拟**\n   *Joon Sung Park、Carolyn Q. Zou、Aaron Shaw、Benjamin Mako Hill、Carrie Cai、Meredith Ringel Morris、Robb Willer、Percy Liang、Michael S. Bernstein。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.10109)]，2024年11月\n\n\n\n\n### 👥 多智能体\n\n#### 🧙 角色\n\n-  **生成式代理人：人类行为的互动模拟物**\n   *Joon Sung Park、Joseph C. O'Brien、Carrie J. Cai、Meredith Ringel Morris、Percy Liang、Michael S. Bernstein。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.03442)]，2023年4月\n\n-  **用于软件开发的沟通型代理人**\n   *Chen Qian、Xin Cong、Wei Liu、Cheng Yang、Weize Chen、Yusheng Su、Yufan Dang、Jiahao Li、Juyuan Xu、Dahai Li、Zhiyuan Liu、Maosong Sun。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.07924)]，2023年7月\n\n-  **Corex：通过多模型协作推动复杂推理的边界**\n   *Qiushi Sun、Zhangyue Yin、Xiang Li、Zhiyong Wu、Xipeng Qiu、Lingpeng Kong。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.00280)]，2023年10月\n\n-  **AVALONBENCH：评估大型语言模型在《阿瓦隆》游戏中的表现。**\n   *乔纳森·莱特、蔡敏、沈晟、胡子牛。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.05036)]，2023年10月。\n\n-  **战争与和平（WarAgent）：基于大型语言模型的世界大战多智能体仿真。**\n   *华文悦、范立洲、李凌瑶、梅凯、季建超、葛英强、莉比·亨菲尔、张永峰。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.17227)]，2023年11月。\n\n-  **利用猜词游戏评估大型语言模型的智能水平。**\n   *梁天、何志伟、黄振泽、王文轩、焦文祥、王睿、杨宇久、涂兆鹏、史树明、王星。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.20499)]，2023年10月。\n\n-  **AntEval：定量评估智能体社交互动的信息性和表达性。**\n   *梁元智、朱林超、杨毅。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.06509)]，2024年1月。\n\n-  **思维交换：通过跨模型通信提升大型语言模型的能力**\n   *尹章越、孙秋实、常程、郭启鹏、戴俊琪、黄宣静、邱锡鹏。* [[摘要](https:\u002F\u002Faclanthology.org\u002F2023.emnlp-main.936\u002F)]，2024年1月。\n\n-  **探索LLM智能体的合作机制：社会心理学视角**\n   *张金田、徐欣、张宁宇、刘瑞博、布莱恩·胡伊、邓淑敏。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.02124)]，2023年10月。\n\n-  **在模拟社交互动上训练社会对齐的语言模型**\n   *刘瑞博、杨睿鑫、贾晨燕、张戈、杨迪一、索鲁什·沃苏吉。* [[摘要](https:\u002F\u002Fopenreview.net\u002Fforum?id=NddKiWtdUm)]，ICLR 2024。\n\n-  **我们在LLM的决策能力方面走到了哪一步？评估LLM在多智能体环境中的博弈能力**\n   *黄振泽、李埃里克·约翰、林满浩、梁天、王文轩、袁友良、焦文祥、王星、涂兆鹏、吕迈克尔·R。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.11807)]，2024年3月。\n\n-  **AgentGroupChat：用于更好激发集体涌现行为的交互式群聊模拟器**\n   *顾周宏、朱晓轩、郭浩然、张琳、蔡音、申昊、陈江杰、叶哲宇、代义飞、高岩、胡耀、冯宏伟、肖阳华* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.13433)]，2024年3月。\n\n-  **这是真实的生活吗？还是仅仅幻想？用LLM模拟社交互动的成功背后的误导**\n   *周旭辉、苏哲、蒂瓦拉约·艾萨佩、金贤宇、马尔滕·萨普。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.05020)]，2024年3月。\n\n-  **EvoAgent：通过进化算法实现多智能体的自动生成**\n   *袁思宇、宋凯涛、陈江杰、谭旭、李东升、杨德清。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.14228)]，2024年6月。\n\n-  **HoLLMwood：通过角色扮演释放大型语言模型在剧本创作中的创造力**\n   *陈静、朱新宇、杨成、石楚凡、奚亚东、张宇翔、王俊杰、蒲嘉树、张荣盛、杨宇久、冯天* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.11683)]，2024年6月。\n\n-  **MathVC：一个由LLM模拟的多角色虚拟教室，用于数学教育**\n   *慕容悦、米夫达尔·维詹、张艺璇、苏珍妮弗、姚子宇。* [[摘要](https:\u002F\u002Fmurongyue.github.io\u002FMathVC.github.io\u002F)]，2024年4月。\n\n-  **人工利维坦：从霍布斯社会契约理论的视角探索LLM智能体的社会演化**\n   *戴戈登、张伟佳、李金汉、杨思琪、奥诺奇·伊贝、拉奥·斯里哈斯、卡埃塔诺·阿瑟、斯拉·米莎。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.14373)]，2024年6月。\n\n-  **对话动作标记：借助多轮规划器引导语言模型进行目标导向的对话**\n   *李肯尼思、王一鸣、维加斯·费尔南达、马丁·瓦滕伯格。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.11978)]，2024年6月。\n\n-  **LLM讨论：通过讨论框架和角色扮演提升大型语言模型的创造力**\n   *卢丽春、陈守仁、裴宗民、余灿鸿、李洪义、孙绍华。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.06373)]，COLM 2024。\n\n-  **多智能体LLM协作中的人格不稳定性：从众、虚构与冒充**\n   *巴尔塔吉·拉赞、赫马蒂安·巴巴克、瓦尔什尼·拉夫·R。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.03862)]，2024年5月。\n\n-  **多专家提示法提升大型语言模型的可靠性、安全性及实用性**\n\n   *杜宣龙、杨玉燕、刘英俊、川口健二、简明彦、陈南希·F。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.00492)]，EMNLP 2024。\n\n-  **项目Sid：迈向AI文明的多智能体模拟**\n   *Altera.AL、安德鲁·安、尼克·贝克、斯蒂芬妮·卡罗尔、尼科·克里斯蒂、曼努埃尔·科尔特斯、阿尔达·德米尔奇、梅丽莎·杜、弗兰基·李、罗淑莹、彼得·Y·王、马修·威洛斯、杨飞彤、杨广宇·罗伯特。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.00114)]，2024年11月。\n\n-  **SocioVerse：一个由LLM智能体和1000万真实用户池驱动的社交模拟世界模型**\n   *张信农、林家宇、牟心怡、杨世悦、刘夏伟、孙利波、吕汉佳、杨以恒、戚卫红、陈岳、李冠英、严玲、胡耀、陈思明、王宇、黄宣静、罗杰波、唐世平、吴利波、周宝华、魏中宇。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.10157)]，2025年4月。\n\n-  **用10亿个智能体模拟地球规模的人类社会**\n   *关浩翔、何继延、范立洋、任真真、何少斌、于欣、陈源、郑淑馨、刘铁燕、刘震。* [[摘要](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2506.12078)]，2025年6月。\n\n\n\n\n\n   *杜宣龙、杨玉燕、刘英俊、川口健二、简明彦、陈南希·F。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.00492)]，EMNLP 2024。\n\n-  **项目Sid：迈向AI文明的多智能体模拟**\n   *Altera.AL、安德鲁·安、尼克·贝克、斯蒂芬妮·卡罗尔、尼科·克里斯蒂、曼努埃尔·科尔特斯、阿尔达·德米尔奇、梅丽莎·杜、弗兰基·李、罗淑莹、彼得·Y·王、马修·威洛斯、杨飞彤、杨广宇·罗伯特。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.00114)]，2024年11月。\n\n-  **SocioVerse：一个由LLM智能体和1000万真实用户池驱动的社交模拟世界模型**\n   *张信农、林家宇、牟心怡、杨世悦、刘夏伟、孙利波、吕汉佳、杨以恒、戚卫红、陈岳、李冠英、严玲、胡耀、陈思明、王宇、黄宣静、罗杰波、唐世平、吴利波、周宝华、魏中宇。* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.10157)]，2025年4月。\n\n-  **用10亿个智能体模拟地球规模的人类社会**\n   *关浩翔、何继延、范立洋、任真真、何少斌、于欣、陈源、郑淑馨、刘铁燕、刘震。* [[摘要](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2506.12078)]，2025年6月。\n\n### 🤖 游戏中的 GUI 代理\n\n- **Lumine：在 3D 开放世界中构建通用代理的开源方案**\n   *Weihao Tan、Xiangyang Li、Yunhao Fang、Heyuan Yao、Shi Yan、Hao Luo、Tenglong Ao、Huihui Li、Hongbin Ren、Bairen Yi、Yujia Qin、Bo An、Libin Liu、Guang Shi.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.08892)]，字节跳动 Seed 2025.11\n\n- **GUI 代理的黎明：与 Claude 3.5 计算机使用相关的初步案例研究**\n   *Siyuan Hu、Mingyu Ouyang、Difei Gao、Mike Zheng Shou.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.10323)]，2024.11\n\n### 🧠 拟人化认知\n\n#### 🎭 性格特质\n\n- **开放性格训练：通过宪法式 AI 塑造 AI 助手的人设**\n   *Sharan Maiya、Henning Bartsch、Nathan Lambert、Evan Hubinger.* [[摘要](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2511.01689)]，2025.11\n\n- **PersonaFuse：一种由人格激活驱动的框架，用于增强人类与 LLM 的交互**\n   *Yixuan Tang、Yi Yang、Ahmed Abbasi.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.07370)]，2025.09\n\n- **人格向量：监控和控制语言模型中的人物特质**\n   *Runjin Chen、Andy Arditi、Henry Sleight、Owain Evans、Jack Lindsey.* [[摘要](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Fpersona-vectors)]，Anthropic，2025.08 \n\n-  **大型语言模型中的性格特质**\n\n   *Mustafa Safdari、Greg Serapio-García、Clément Crepy、Stephen Fitz、Peter Romero、Luning Sun、Marwa Abdulhai、Aleksandra Faust、Maja Matarić.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.00184)]，2023.7\n\n-  **白盒语言模型的性格估计。**\n\n   *Saketh Reddy Karra、Son The Nguyen、Theja Tulabandhula.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.12000)]，2022.4\n\n-  **PersonaLLM：探究 GPT-3.5 表达性格特质及性别差异的能力。**\n    \n    *Hang Jiang、Xiajie Zhang、Xubo Cao、Jad Kabbara.*，[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.02547)]，2023.5\n\n-  **GPT-3 是否表现出精神病态？从心理学角度评估大型语言模型。**\n\n   *Xingxuan Li、Yutong Li、Shafiq Joty、Linlin Liu、Fei Huang、Lin Qiu、Lidong Bing.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.10529)]，2022.12\n\n-  **预训练语言模型中性格的评估与诱导。**\n   *Guangyuan Jiang、Manjie Xu、Song-Chun Zhu、Wenjuan Han、Chi Zhang、Yixin Zhu.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.07550)]，2022.6\n\n-  **重新审视大型语言模型上心理量表的可靠性（此前：ChatGPT 是 ENFJ，Bard 是 ISTJ：大型语言模型性格的实证研究）。**\n   *Jen-tse Huang、Wenxuan Wang、Man Ho Lam、Eric John Li、Wenxiang Jiao、Michael R. Lyu.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.19926)]，2023.5\n\n-  **LLM 是否具有性格？将 MBTI 测试作为大型语言模型的绝佳评估工具。**\n\n   *Keyu Pan、Yawen Zeng.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.16180)]，2023.7\n\n-  **ChatGPT 能否评估人类性格？一个通用的评估框架。**\n\n   *Haocong Rao、Cyril Leung、Chunyan Miao.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.01248)]，2023.3\n\n-  **GPT-3 是谁？对性格、价值观和人口统计特征的探索。**\n\n   *Marilù Miotto、Nicola Rossberg、Bennett Kleinberg.*，[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.14338)]，2022.9\n\n-  **为 LLM 编辑性格**\n    \n   *Shengyu Mao、Ningyu Zhang、Xiaohan Wang、Mengru Wang、Yunzhi Yao、Yong Jiang、Pengjun Xie、Fei Huang、Huajun Chen.*，[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.02168)]，2023.10\n\n\n-  **利用电影剧本预测虚构角色的 MBTI 性格**\n    \n    *Yisi Sang、Xiangyang Mou、Mo Yu、Dakuo Wang、Jing Li、Jeffrey Stanton.*，[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.10994)]，EMNLP 2022\n\n-  **释放大型语言模型中的认知协同效应：通过多个人设自我协作的任务解决代理**\n    \n    *Zhenhailong Wang、Shaoguang Mao、Wenshan Wu、Tao Ge、Furu Wei、Heng Ji.*，[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.05300)]，2023.7\n   \n-  **开放模型，封闭心智？关于开放型大型语言模型模仿人类性格的能力的探讨**\n\n   *Lucio La Cava、Davide Costa、Andrea Tagarelli.*，[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.07115)]，2024.01\n\n- **1,000 人的生成式代理模拟**\n   *Joon Sung Park、Carolyn Q. Zou、Aaron Shaw、Benjamin Mako Hill、Carrie Cai、Meredith Ringel Morris、Robb Willer、Percy Liang、Michael S. Bernstein.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.10109)]，2024.11\n\n\n#### 👥 社会智能与心理理论\n\n- **SI-Bench：在人与人对话中基准测试大型语言模型的社会智能**\n   *Shuai Huang、Wenxuan Zhao、Jun Gao.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.23182)]，2025.10\n\n-  **心理理论可能已在大型语言模型中自发出现**\n\n   *Michal Kosinski.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2302.02083)]，2023.2\n\n-  **SOTOPIA：语言代理社会智能的交互式评估**\n\n   *Xuhui Zhou、Hao Zhu、Leena Mathur、Ruohong Zhang、Haofei Yu、Zhengyang Qi、Louis-Philippe Morency、Yonatan Bisk、Daniel Fried、Graham Neubig、Maarten Sap.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.11667)]，2023.10\n\n- **SOTOPIA-π：社交智能语言代理的交互式学习**\n\n   *Ruiyi Wang、Haofei Yu、Wenxin Zhang、Zhengyang Qi、Maarten Sap、Graham Neubig、Yonatan Bisk、Hao Zhu.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.08715)]，2024.03\n\n- **AgentSense：通过交互场景基准测试语言代理的社会智能**\n   *Xinyi Mou、Jingcong Liang、Jiayu Lin、Xinnong Zhang、Xiawei Liu、Shiyue Yang、Rong Ye、Lei Chen、Haoyu Kuang、Xuanjing Huang、Zhongyu Wei* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.19346)]，2024.10\n\n- **迈向在行动层面客观基准测试语言代理的社会智能**\n   *Chenxu Wang、Bin Dai、Huaping Liu、Baoyuan Wang.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.05337)]，2024.04\n\n-  **OpenToM：全面基准测试大型语言模型心理理论推理能力的方案。**\n   *Hainiu Xu、Runcong Zhao、Lixing Zhu、Jinhua Du、Yulan He.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.06044)]，2024.02。\n\n-  **这是真实的生活吗？还是只是幻想？用 LLM 模拟社交互动的误导性成功**\n   *Xuhui Zhou、Zhe Su、Tiwalayo Eisape、Hyunwoo Kim、Maarten Sap* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.05020)]，2024.03\n\n- **SocialEval：评估大型语言模型的社会智能**\n  *Jinfeng Zhou、Yuxuan Chen、Yihan Shi、Xuanming Zhang、Leqi Lei、Yi Feng、Zexuan Xiong、Miao Yan、Xunzhi Wang、Yaru Cao、Jianing Yin、Shuai Wang、Quanyu Dai、Zhenhua Dong、Hongning Wang、Minlie Huang.* [[摘要](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2506.00900)]，2025.06\n\n- **Sotopia-RL：社交智能的奖励设计**\n  *Haofei Yu、Zhengyang Qi、Yining Zhao、Kolby Nottingham、Keyang Xuan、Bodhisattwa Prasad Majumder、Hao Zhu、Paul Pu Liang、Jiaxuan You.* [[摘要](https:\u002F\u002Fwww.arxiv.org\u002Fabs\u002F2508.03905)]，2025.08\n\n#### 🧠 其他 \u003C!---others-1-->\n\n-  **与大型语言模型的角色扮演**\n   *默里·沙纳汉、凯尔·麦克唐奈尔和拉里娅·雷诺兹.*[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.16367)], 2023年5月\n\n-  **人工智能中的意识：来自意识科学的洞见。**\n   *帕特里克·巴特林、罗伯特·朗、埃里克·埃尔莫兹尼诺、约书亚·本吉奥、乔纳森·伯奇、阿克塞尔·康斯坦特、乔治·迪恩、斯蒂芬·M·弗莱明、克里斯·弗里斯、徐继、梁内良、科林·克莱因、格蕾丝·林赛、马蒂亚斯·米歇尔、利亚德·穆德里克、梅根·A·K·彼得斯、埃里克·施维茨盖贝尔、乔纳森·西蒙、鲁芬·范鲁伦.*[[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.08708)], 2023年8月\n\n-  **ChatGPT的自我认知与政治偏见。**\n\n   *热罗姆·鲁蒂诺夫斯基、斯文·弗兰克、扬·恩登迪克、伊娜·多尔穆特、马库斯·保利*, 2023年4月。标签：人格特质、政治偏见、黑暗人格特质。\n\n-  **对话式AI的政治意识形态：关于ChatGPT亲环境、左翼自由主义取向的汇聚证据。**\n\n   *约亨·哈特曼、贾斯珀·施文佐、马克西米利安·维特*, 2023年1月。标签：政治偏见。\n\n-  **在大型语言模型中诱发焦虑会增加探索性和偏见。**\n\n   *朱利安·科达-福尔诺、克里斯汀·维特、阿克谢·K·贾加迪什、马塞尔·宾茨、泽内普·阿卡塔、埃里克·舒尔茨*, 2023年4月。标签：焦虑。\n\n-  **从社会心理学视角探讨LLM智能体的合作机制。**\n\n   *张金田、许欣、邓淑敏*, 2023年10月。标签：社会心理学。\n\n- **REALTALK：用于长期对话的21天真实世界数据集**\n   *李东浩、阿迪亚莎·马哈拉纳、杰伊·普贾拉、任翔、弗朗切斯科·巴比耶里.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13270)], 2025年2月\n\n\n\n### 🌐 其他 \u003C!---others-2-->\n\n#### 📂 资源\n\n- **模型：CoSER-8B、CoSER-70B**\n   *王新涛等.* [[huggingface](https:\u002F\u002Fgithub.com\u002FNeph0s\u002FCOSER)]\n\n- **模型：Deepsex**\n   *ValueFX9507.* [[huggingface](https:\u002F\u002Fhuggingface.co\u002FValueFX9507\u002FTifa-Deepsex-14b-CoT-GGUF-Q4)]\n\n- **基准测试：RPBench排行榜**\n   *Boson AI.* [[排行榜](https:\u002F\u002Fboson.ai\u002Frpbench\u002F)]\n\n- **基准测试：Fiction.liveBench**\n   *Fiction.live.* [[基准测试](https:\u002F\u002Ffiction.live\u002Fstories\u002FFiction-liveBench-Mar-14-2025\u002FoQdzQvKHw8JyXbN87\u002Fhome)]\n\n- **模型：Higgs-Llama-3-70B**\n   *Boson AI.* [[huggingface](https:\u002F\u002Fhuggingface.co\u002Fbosonai\u002FHiggs-Llama-3-70B)]\n\n#### 📰 博客\n\n- **[[情绪概念及其在大型语言模型中的作用](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Femotion-concepts-function)]**, Anthropic博客, 2026年4月\n\n- **[[助手轴：定位并稳定大型语言模型的性格特征](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Fassistant-axis)]**, Anthropic博客, 2026年1月\n\n- **[[人格向量：监控和控制语言模型中的性格特征](https:\u002F\u002Fwww.anthropic.com\u002Fresearch\u002Fpersona-vectors)]**, Anthropic博客, 2025年8月\n- \n- **[[Character.ai赛道的真相 (The Truth About the Character.ai Track)](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002FH2KNDGRNHktHiQc3sayFsA)]**, 2024年8月\n\n- **[[李沐：创业一年，人间三年！(Mu Li (Boson AI): One Year of Entrepreneurship, Three Years in the Human World!)](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002F2lbCMo64-nU5yRz1cLQxYA)]**, 2024年8月\n\n- **[[盘点ACL 2024角色扮演方向都发了啥 (A Review of Publications in Role-Playing Direction at ACL 2024)](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002FE5qp5YPYPVaLM07OumDTRw)]**, 2024年7月\n\n- **[[角色扮演大模型的碎碎念 (Ramblings on Large Language Models for Role-Playing)](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002FyoM-srJYGGfyd1VXirg_Hg)]**, 2024年3月\n\n- **[[聊一聊做角色扮演大模型的经验 (Discussing Experiences in Developing Role-Playing Large Language Models)](https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F3641608640)]**, 2024年10月\n\n- **[[角色扮演大模型技术分享 (Technical Insights on Role-Playing Large Language Models)](https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F685823865)]**, 2024年3月\n\n- **[[角色扮演大模型技术分享2—超拟人模型的困境 (Technical Insights on Role-Playing Large Language Models 2 - The Dilemma of Super-Anthropomorphic Models)](https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F719135803)]**, 2024年9月\n\n- **[[角色扮演大模型技术分享3—提升拟人能力及数据合成流水线 (Technical Insights on Role-Playing Large Language Models 3 - Improving Anthropomorphic Capabilities & Data Synthesis Pipeline)](https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F719772276)]**, 2024年9月\n\n\n#### 📄 论文\n\n- **一千零一对：面向长上下文语言模型的“新颖”挑战**\n   *马尔泽纳·卡尔平斯卡、凯瑟琳·泰、凯尔·洛、坦雅·戈亚尔、莫希特·艾耶尔.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.16264)], 2024年6月\n\n- **NovelQA：对超过20万标记文档的问答进行基准测试**\n   *王存祥、宁若曦、潘博奇、吴通辉、郭启鹏、邓成、鲍广生、胡湘坤、张正、王倩、张悦.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.12766)], 2024年3月\n\n\n- **视觉-角色扮演：通过角色扮演图像角色对多模态大型语言模型发起的通用越狱攻击**\n   *马思远、罗伟迪、王宇、刘晓耕.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.20773)], 2024年6月\n\n- **AI猫叙述者：设计一款用于探索与猫共享的世界及社交连接的AI工具**\n   *赖振驰、黄珍英、梁荣慧.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.06192)], 2024年6月\n\n- **THEANINE：通过时间线增强的回答生成重新审视长期对话中的内存管理**\n   *金瑞贤、翁凯子云、权泰润、金南永、姜琴敏、裴成贤、赵耀韩、黄承元、李东河、吕珍英.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.10996)], 2024年6月\n\n- **ESC-Eval：评估大型语言模型中的情绪支持对话**\n   *赵海泉、李凌宇、陈世松、孔淑琪、王佳安、黄可欣、顾天乐、王义旭、梁丹丹、李志旭、滕岩、肖阳华、王迎春.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.14952)], 2024年6月\n\n- **PATIENT-Ψ：利用大型语言模型模拟患者以培训心理健康专业人员**\n   *王睿怡、斯蒂芬妮·米拉尼、杰米·C·邱、支嘉茵、肖恩·M·伊克、特拉维斯·拉布鲁姆、塞缪尔·M·墨菲、内夫·琼斯、凯特·哈迪、沈宏、方飞、陈智宇.* [[摘要](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.19660)], 2024年5月\n\n- **CharacterMeet：通过与LLM驱动的聊天机器人化身对话，支持创意写作者完成整个故事角色构建过程**\n   *秦华轩、晋山、高泽、范明明、惠潘.* [[摘要](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3613904.3642105)], CHI 2024。\n\n- **Stephanie：用于模仿社交对话中人类互动的逐步对话**\n   *杨浩、陆鸿远、曾新华、刘洋、张翔、杨浩然、张雨萌、魏依冉、韦怀.* [[摘要](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2407.04093)], 2024年7月\n\n\n---\n\n## 💡 贡献\n\n### 参与本仓库的贡献\n\n🤲\" **加入我们，一起完善这个仓库吧！** 如果你发现了我们可能遗漏的重要作品，请随时添加。每一次贡献都很重要！   \"\n\n### 贡献者\n\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FNeph0s\u002Fawesome-llm-role-playing-with-persona\u002Fgraphs\u002Fcontributors\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FNeph0s_awesome-llm-role-playing-with-persona_readme_6d8b2ac02506.png\" \u002F>\n\u003C\u002Fa>","# awesome-llm-role-playing-with-persona 快速上手指南\n\n> **注意**：本项目是一个**论文与资源汇总列表（Awesome List）**，而非一个可直接安装运行的软件库或框架。它主要收录了关于“大语言模型角色扮演（Role-Playing）”和“人设（Persona）”相关的学术论文、基准测试及开源项目链接。\n>\n> 因此，本指南将指导你如何**获取资源**以及**如何使用列表中推荐的典型工具**进行快速体验。\n\n## 环境准备\n\n由于本项目本身不包含代码执行逻辑，若要运行列表中推荐的具体角色扮演模型（如 Character-LLM, ChatHaruhi 等），通常需要以下基础环境：\n\n*   **操作系统**: Linux (推荐 Ubuntu 20.04+), macOS 或 Windows (WSL2)\n*   **Python**: 3.8 或更高版本\n*   **依赖管理**: pip 或 conda\n*   **硬件要求** (视具体模型而定):\n    *   推理小型模型：至少 8GB RAM\n    *   推理\u002F微调大型模型：建议配备 NVIDIA GPU (显存 16GB+)\n*   **前置依赖**:\n    ```bash\n    pip install torch torchvision torchaudio --index-url https:\u002F\u002Fdownload.pytorch.org\u002Fwhl\u002Fcu118\n    # 国内用户推荐使用清华源加速 PyTorch 安装\n    # pip install torch torchvision torchaudio --index-url https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n    ```\n\n## 安装步骤\n\n本项目无需传统意义上的“安装”，只需克隆仓库以获取最新的论文列表和资源索引。\n\n1.  **克隆仓库**:\n    ```bash\n    git clone https:\u002F\u002Fgithub.com\u002FNeph0s\u002Fawesome-llm-role-playing-with-persona.git\n    cd awesome-llm-role-playing-with-persona\n    ```\n\n2.  **获取核心资源**:\n    浏览仓库中的 `README.md` 文件，找到你感兴趣的特定项目（例如 `Character-LLM` 或 `ChatHaruhi`），点击链接跳转至其独立仓库进行单独安装。\n\n    *示例：若想尝试列表中提到的 **Character-LLM**:*\n    ```bash\n    git clone https:\u002F\u002Fgithub.com\u002Fscutcyr\u002FCharacter-LLM.git\n    cd Character-LLM\n    pip install -r requirements.txt\n    ```\n\n## 基本使用\n\n由于这是一个资源列表，\"使用\"通常指参考列表中的论文实现代码，或利用通用的角色扮演 Prompt 技巧。以下是基于列表中常见方法的**最小化使用示例**。\n\n### 方式一：使用通用 LLM 进行简单角色扮演 (Prompt Engineering)\n\n无需安装额外模型，直接使用支持 System Prompt 的 LLM API 或本地模型。\n\n```python\nfrom openai import OpenAI\n\n# 初始化客户端 (以兼容 OpenAI 格式的本地模型为例)\nclient = OpenAI(\n    base_url=\"http:\u002F\u002Flocalhost:8000\u002Fv1\", \n    api_key=\"not-needed\"\n)\n\n# 定义人设 (Persona)\nsystem_prompt = \"\"\"\n你现在的角色是“林黛玉”。\n性格特点：多愁善感、才华横溢、说话委婉但带刺。\n背景：红楼梦中的女主角，居住在贾府大观园。\n请始终保持这个角色设定与我对话，不要跳出角色。\n\"\"\"\n\nuser_input = \"今天天气真好，我们去放风筝吧？\"\n\nresponse = client.chat.completions.create(\n    model=\"character-model\", # 替换为你加载的角色模型名称\n    messages=[\n        {\"role\": \"system\", \"content\": system_prompt},\n        {\"role\": \"user\", \"content\": user_input}\n    ],\n    temperature=0.7, # 较高的温度有助于发挥角色创造性\n    max_tokens=200\n)\n\nprint(response.choices[0].message.content)\n```\n\n### 方式二：运行列表中推荐的开源项目 (以 Character-LLM 为例)\n\n如果你克隆了列表中具体的项目仓库，通常可以通过脚本加载特定角色配置文件。\n\n```bash\n# 进入具体项目目录 (假设已安装 Character-LLM)\ncd Character-LLM\n\n# 运行推理脚本，指定角色配置文件\n# 注意：具体命令需参照该子项目的 README，以下为典型示例\npython inference.py \\\n    --model_path .\u002Fmodels\u002Fllama-7b \\\n    --persona_config .\u002Fcharacters\u002Flin_daiyu.json \\\n    --input \"姑娘今日为何如此伤感？\"\n```\n\n### 核心资源索引\n\n在使用时，请参考本仓库 `README` 中的分类查找所需技术：\n*   **🧙 Characters**: 虚构人物、名人扮演 (如 ChatHaruhi, Character-LLM)\n*   **👨‍👩‍👧‍👦 Demographics**: 人口统计学特征模拟\n*   **🤖 Personalization**: 个性化代理构建\n*   **📂 Resources**: 数据集与评估基准 (如 CharacterEval, RoleLLM)\n\n建议定期 `git pull` 更新此仓库，以获取最新的研究成果（如 2024-2026 年的最新论文）。","某游戏开发团队正在为一款历史题材的互动叙事游戏构建 NPC 对话系统，需要让秦始皇、李白等历史人物展现出符合史实的性格与说话风格。\n\n### 没有 awesome-llm-role-playing-with-persona 时\n- 开发人员需在海量论文中盲目搜索，难以区分哪些研究真正专注于“角色一致性”而非通用的多智能体协作。\n- 自行设计的提示词（Prompt）往往导致角色“出戏”，例如李白开始谈论现代科技，缺乏权威的方法论来约束模型行为。\n- 团队重复造轮子，花费数周时间复现基础的角色扮演效果，却错过了如 ChatHaruhi 或 RoleLLM 等成熟的开源基准与优化方案。\n- 缺乏对角色心理理论（Theory-of-Mind）和人格特质的系统性参考，导致 NPC 对话肤浅，无法通过图灵测试般的沉浸感评估。\n\n### 使用 awesome-llm-role-playing-with-persona 后\n- 团队直接利用该仓库 curated 的资源列表，快速定位到 TMLR 接收的最新综述及 ICML 前沿论文，明确了技术选型路径。\n- 基于仓库推荐的 CoSER 和 RoleLLM 等框架，迅速部署了高精度的角色模拟管线，确保李白只谈诗词酒赋，秦始皇尽显帝王威严。\n- 复用社区已验证的评测基准（Benchmark），将原本数周的调优周期缩短至几天，并直接集成了先进的个性化生成算法。\n- 借鉴关于社会智能与人格特质的研究成果，为 NPC 注入了深层的认知逻辑，使玩家能体验到具有情感深度和历史厚度的互动剧情。\n\nawesome-llm-role-playing-with-persona 通过聚合顶尖学术资源与实战框架，将角色扮演的研发从“盲目试错”转变为“站在巨人肩膀上”的高效创新。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FNeph0s_awesome-llm-role-playing-with-persona_d09eaf5f.png","Neph0s","Xintao Wang","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FNeph0s_9e85a7d3.png","PhD student from Fudan University. Interested in Natural Language Processing, especially role-playing AI.","Fudan University","Shanghai",null,"https:\u002F\u002Fgithub.com\u002FNeph0s",1007,52,"2026-04-11T21:46:28","","未说明",{"notes":87,"python":85,"dependencies":88},"该仓库是一个关于“角色扮演语言模型”的论文和资源清单（Awesome List），并非可执行的软件工具或代码库。因此，它不包含具体的运行环境需求、依赖库或安装说明。用户仅需浏览网页或阅读列出的论文链接即可使用本资源。",[],[13,15,35,14],[91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107],"agent","ai","aigc","awesome","awesome-list","character","chatgpt","conversational-ai","deep-learning","large-language-models","llm","natural-language-processing","nlp","paper-list","persona","role-playing","survey","2026-03-27T02:49:30.150509","2026-04-18T14:35:21.295414",[],[]]