Conventions are crucial for strong performance in cooperative multi-agent games, because they allow players to coordinate on a shared strategy without explicit communication. Unfortunately, standard multi-agent reinforcement learning techniques, such as self-play, converge to conventions that are arbitrary and non-diverse, leading to poor generalization when interacting with new partners. In this work, we present a technique for generating diverse conventions by (1) maximizing their rewards during self-play, while (2) minimizing their rewards when playing with previously discovered conventions (cross-play), stimulating conventions to be semantically different. To ensure that learned policies act in good faith despite the adversarial optimization of cross-play, we introduce \emph{mixed-play}, where an initial state is randomly generated by sampling self-play and cross-play transitions and the player learns to maximize the self-play reward from this initial state. We analyze the benefits of our technique on various multi-agent collaborative games, including Overcooked, and find that our technique can adapt to the conventions of humans, surpassing human-level performance when paired with real users.
OMAC: A Holistic Optimization Framework for LLM-Based Multi-Agent Collaboration
Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, United States(得克萨斯大学奥斯汀分校电子与计算机工程系);Intuit AI Research, Mountain View, United States(Intuit AI研究)
ATOM: Instantiating Budget-Controllable Multi-Agent Collaboration via Nucleus-Electron Hierarchy
Zhejiang University(浙江大学);Ningbo Global Innovation Center, Zhejiang University(浙江大学宁波全球创新中心);Zhejiang Key Laboratory of Digital-Intelligence Service Technology(浙江省数字智能服务技术重点实验室);Xidian University(西安电子科技大学)
Multi-Agent Collaboration via Evolving Orchestration
Tsinghua University(清华大学);Shanghai Jiao Tong University(上海交通大学);Beijing University of Posts and Telecommunications(北京邮电大学);Siemens(西门子);Tencent Robotics X(腾讯机器人研究院)
MedAgentBoard: Benchmarking Multi-Agent Collaboration with Conventional Methods for Diverse Medical Tasks
National Engineering Research Center for Software Engineering, Peking University(北京大学国家软件工程研究中心);School of Computing and Data Science, The University of Hong Kong(香港大学计算与数据科学学院);Centre for Medical Informatics, The University of Edinburgh(爱丁堡大学医学信息学中心);ETH Zurich(苏黎世联邦理工学院);Health Data Research UK(英国健康数据研究所);Key Laboratory of High Confidence Software Technologies, Ministry of Education(教育部高可信软件技术重点实验室)
CoHuB: A Simulation Benchmark for Multi-Humanoid Collaboration
Yonsei University(延世大学);Gwangju Institute of Science and Technology(光州科学技术院);University of California, Berkeley(加州大学伯克利分校);Seoul National University(首尔大学)
CollabBench: Benchmarking and Unleashing Collaborative Ability of LLMs with Diverse Players via Proactive Engagement
Shanghai Institute of AI for Education(上海人工智能教育研究院);School of Computer Science(计算机科学学院);East China Normal University(东华大学);Tencent Inc.(腾讯公司);Shanghai Innovation Institute(上海创新研究院)
ProtAgents: Protein discovery via large language model multi-agent collaborations combining physics and machine learning
Massachusetts Institute of Technology(麻省理工学院);Laboratory for Atomistic and Molecular Mechanics (LAMM)(原子与分子力学实验室);Center for Computational Science and Engineering(计算科学与工程中心);Schwarzman College of Computing(施瓦茨曼计算学院)
Empowering Edge Agents: A Systematic Analysis of Edge-Cloud Collaboration in LLM Agent Execution
School of Computer Science and Engineering, Northeastern University(东北大学计算机科学与工程学院);Department of Computer Science and Technology, Tsinghua University(清华大学计算机科学与技术系);China Mobile Online Services Co., Ltd.(中移在线服务有限公司)