Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration
Zhenhailong Wang1, Shaoguang Mao2, Wenshan Wu2, Tao Ge2, Furu Wei2, Heng Ji1 1University of Illinois Urbana-Champaign, 2Microsoft Research Asia Work was done when interning at Microsoft Research Asia. Corresponding author.
Abstract
Human intelligence thrives on cognitive synergy, where collaboration among different minds yield superior outcomes compared to isolated individuals. In this work, we propose Solo Performance Prompting (SPP), which transforms a single LLM into a cognitive synergist by engaging in multi-turn self-collaboration with multiple personas. A cognitive synergist is an intelligent agent that collaboratively combines multiple minds’ strengths and knowledge to enhance problem-solving in complex tasks. By dynamically identifying and simulating different personas based on task inputs, SPP unleashes the potential of cognitive synergy in LLMs. Our in-depth analysis shows that assigning multiple fine-grained personas in LLMs improves problem-solving abilities compared to using a single or fixed number of personas. We evaluate SPP on three challenging tasks: Trivia Creative Writing, Codenames Collaborative, and Logic Grid Puzzle, encompassing both knowledge-intensive and reasoning-intensive types. Unlike previous works, such as Chain-of-Thought, that solely enhance the reasoning abilities in LLMs, experimental results demonstrate that SPP effectively reduces factual hallucination, and maintains stro
中文速览
大语言模型（LLM）在知识密集型任务中容易"一本正经地胡说八道"（事实幻觉），在推理密集型任务中也难以像人类团队那样发挥协同优势——这篇论文针对这两个痛点，提出了"独角表演提示"（Solo Performance Prompting，SPP）方法。SPP让单个LLM在一次对话中自动扮演多个角色（如领域专家、目标受众等），通过头脑风暴、相互反馈、迭代修改来模拟多智能体协作，整个过程无需额外模型、检索系统或微调。在三项挑战性任务——知识密集的"百科创意写作"、需要心智理论的"代号牌合作"以及纯推理的"逻辑网格谜题"——上，SPP在GPT-4上显著优于标准提示、思维链和自我精炼等基线，既减少了事实幻觉，又保持了强推理能力。值得注意的是，这种"认知协同"能力只在GPT-4中涌现，GPT-3.5和Llama2-13b上并不出现，与儿童要到2至3岁才开始角色扮演的发育规律形成有趣类比，揭示了LLM能力涌现的新维度。
原文 arXiv:2307.05300;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2307.05300v4