学会打破：多智能体辩论系统中的知识增强推理
Learning to Break: Knowledge-Enhanced Reasoning in Multi-Agent Debate System
AI总结：
针对多智能体辩论系统中因认知孤岛阻碍寻优的问题，提出带共享检索知识库与自适应知识选择的MADKE框架，在六数据集上达SOTA，开源LLM搭载后性能超GPT-4。
摘要：
多智能体辩论系统（Multi-agent debate system，MAD）模仿人类求真讨论的过程，旨在对齐不同智能体的正确认知以寻求最优解。由于各智能体的知识背景有限且存在差异（即认知孤岛），让各类智能体形成正确且高度一致的认知颇具挑战性，这会阻碍最优解的探寻。为应对这一挑战，我们提出了一种新颖的知识增强多智能体辩论框架（Multi-Agent Debate with Knowledge-Enhanced framework，MADKE），以推动系统找到解决方案。首先，我们在辩论过程中引入共享检索知识库，以解决知识背景有限且存在差异的问题。随后，我们提出了一种自适应知识选择方法，以保证知识的准确性和个性化。该方法允许智能体根据自身需求，在每一轮对话中选择是否使用外部知识。我们在六个数据集上的实验结果表明，与现有的单智能体和多智能体方法相比，我们的方法取得了当前最优（state-of-the-art）结果。进一步分析显示，检索知识的引入能够帮助智能体在辩论过程中打破认知孤岛，有效提升模型的一致性和正确性。此外，采用Qwen1.5-72B-Chat的MADKE在六个数据集上的平均表现超过GPT-4 1.26个百分点，这验证了我们的方法能够帮助开源大语言模型（LLM）达到甚至超越GPT-4的性能。我们的代码可在https://github.com/FutureForMe/MADKE获取。
英文摘要：
Multi-agent debate system (MAD) imitating the process of human discussion in pursuit of truth, aims to align the correct cognition of different agents for the optimal solution. It is challenging to make various agents perform right and highly consistent cognition due to their limited and different knowledge backgrounds (i.e., cognitive islands), which hinders the search for the optimal solution. To address the challenge, we propose a novel \underline{M}ulti-\underline{A}gent \underline{D}ebate with \underline{K}nowledge-\underline{E}nhanced framework (\textbf{MADKE}) to promote the system to find the solution. First, we involve a shared retrieval knowledge pool in the debate process to solve the problem of limited and different knowledge backgrounds. Then, we propose an adaptive knowledge selection method to guarantee the accuracy and personalization of knowledge. This method allows agents to choose whether to use external knowledge in each conversation round according to their own needs. Our experimental results on six datasets show that our method achieves state-of-the-art results compared to existing single-agent and multi-agent methods. Further analysis reveals that the introduction of retrieval knowledge can help the agent to break cognitive islands in the debate process and effectively improve the consistency and correctness of the model. Moreover, MADKE using Qwen1.5-72B-Chat surpasses GPT-4 by +1.26\% on average in six datasets, which validates that our method can help open-source LLMs achieve or even surpass the performance of GPT-4. Our code is available at \url{https://github.com/FutureForMe/MADKE}.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
