Large language models, employed as multiple agents that interact and collaborate with each other, have excelled at solving complex tasks. The agents are programmed with prompts that declare their functionality, along with the topologies that orchestrate interactions across agents. Designing prompts and topologies for multi-agent systems (MAS) is inherently complex. To automate the entire design process, we first conduct an in-depth analysis of the design space aiming to understand the factors behind building effective MAS. We reveal that prompts together with topologies play critical roles in enabling more effective MAS design. Based on the insights, we propose Multi-Agent System Search (MASS), a MAS optimization framework that efficiently exploits the complex MAS design space by interleaving its optimization stages, from local to global, from prompts to topologies, over three stages: 1) block-level (local) prompt optimization; 2) workflow topology optimization; 3) workflow-level (global) prompt optimization, where each stage is conditioned on the iteratively optimized prompts/topologies from former stages. We show that MASS-optimized multi-agent systems outperform a spectrum of existing alternatives by a substantial margin. Based on the MASS-found systems, we finally propose design principles behind building effective multi-agent systems.
TCP-MCP: Landscape-Guided Co-Evolution of Prompts and Communication Topologies for Multi-Agent Systems
National Institute of Metrology, China(中国计量科学研究院);University of California, Berkeley(加州大学伯克利分校);Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(中国科学院深圳先进技术研究院);Nanjing University of Chinese Medicine(南京中医药大学);WEEX Exchange(WEEX交易所);National University of Singapore(新加坡国立大学);Wuhan University(武汉大学);Peking University(北京大学)
RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learning
Carnegie Mellon University(卡内基梅隆大学);UC San Diego(加州大学圣地亚哥分校);MIT(麻省理工学院);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学);Petuum Inc.(Petuum公司)
Optimizing Soft Prompt Tuning via Structural Evolution
School of Information and Software Engineering, University of Electronic Science and Technology of China(电子科技大学信息与软件工程学院);School of Computer Science and Engineering, University of Electronic Science and Technology of China(电子科技大学计算机科学与工程学院);College of Professional and Continuing Education, The Hong Kong Polytechnic University(香港理工大学专业及继续教育学院);School of Robotics and Advanced Manufacture, Harbin Institute of Technology(哈尔滨工业大学机器人与先进制造学院);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);School of Computer Science and Technology, Tongji University(同济大学计算机科学与技术学院)
SPRIG: Improving Large Language Model Performance by System Prompt Optimization
University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);University of Michigan(密歇根大学);University of Illinois Chicago(伊利诺伊大学芝加哥分校);LG AI Research(LG AI 研究院)
AutoMedPrompt: A New Framework for Optimizing LLM Medical Prompts Using Textual Gradients
Keck Data Science Institute, Pepperdine University(佩珀代因大学凯克数据科学研究所);Department of Medicine, University of California, Los Angeles(加利福尼亚大学洛杉矶分校医学院);Biological Sciences Division, University of Chicago(芝加哥大学生物科学部);Brain Research Institute, University of California, Los Angeles(加利福尼亚大学洛杉矶分校脑研究所)
MetaGen: Self-Evolving Roles and Topologies for Multi-Agent LLM Reasoning
School of Artificial Intelligence, Jilin University(吉林大学人工智能学院);Department of Computer Science and Engineering, Southern University of Science and Technology(南方科技大学计算机科学与工程系);School of Urban Planning and Design, Peking University(北京大学城市规划与设计学院)
Optimizing Social Media Annotation of HPV Vaccine Skepticism and Misinformation Using Large Language Models: An Experimental Evaluation of In-Context Learning and Fine-Tuning Stance Detection Across Multiple Models
School of Journalism and Mass Communication, University of Wisconsin–Madison(威斯康星大学麦迪逊分校新闻与大众传播学院);Department of Electrical & Computer Engineering, University of Wisconsin–Madison(威斯康星大学麦迪逊分校电气与计算机工程系);Department of Computer Science, University of Wisconsin–Madison(威斯康星大学麦迪逊分校计算机科学系);Department of Psychology, University of Wisconsin–Madison(威斯康星大学麦迪逊分校心理学系);School of Interactive Computing, Georgia Institute of Technology(佐治亚理工学院交互计算学院);Institute of People and Technology, Georgia Institute of Technology(佐治亚理工学院人与技术研究所)