arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.06307cs.AI
收藏

信念校准的多智能体共识寻求用于复杂自然语言处理任务

Belief-Calibrated Multi-Agent Consensus Seeking for Complex NLP Tasks

Shandong University(山东大学);VU Amsterdam(阿姆斯特丹自由大学);Leiden University(莱顿大学)

Wentao Deng, Jiahuan Pei, Zhiwei Xu, Zhaochun Ren, Zhumin Chen, Pengjie Ren

AI总结:

针对多智能体共识寻求中忽视内部信念矛盾与协作对象选择不当的问题,提出BCCS框架,通过选择最优协作者并校准共识判断,在MATH和MMLU上分别提升2.23%和3.95%准确率。

摘要:

多智能体系统(MAS)通过多个智能体之间的协作来增强其解决复杂自然语言处理(NLP)任务的能力,其中共识寻求是一个基本机制。然而,现有的共识寻求方法通常依赖投票机制来判断共识,忽视了系统内部信念中的矛盾,这些矛盾会破坏共识的稳定性。此外,这些方法通常涉及智能体通过与每个其他智能体不加区分地协作来更新其结果。这种统一交互未能为每个智能体识别最佳协作者,阻碍了稳定共识的出现。为了解决这些挑战,我们提供了一个理论框架,用于选择最大化共识稳定性的最优协作者。基于这些定理,我们提出了信念校准的共识寻求(BCCS)框架,通过选择最优协作者并根据系统内部信念校准共识判断来促进稳定共识。在MATH和MMLU基准数据集上的实验结果表明,所提出的BCCS框架在具有挑战性的任务上分别比现有最佳结果高出2.23%和3.95%的准确率。我们的代码和数据可在https://github.com/dengwentao99/BCCS获取。

英文摘要:

A multi-agent system (MAS) enhances its capacity to solve complex natural language processing (NLP) tasks through collaboration among multiple agents, where consensus-seeking serves as a fundamental mechanism. However, existing consensus-seeking approaches typically rely on voting mechanisms to judge consensus, overlooking contradictions in system-internal beliefs that destabilize the consensus. Moreover, these methods often involve agents updating their results through indiscriminate collaboration with every other agent. Such uniform interaction fails to identify the optimal collaborators for each agent, hindering the emergence of a stable consensus. To address these challenges, we provide a theoretical framework for selecting optimal collaborators that maximize consensus stability. Based on the theorems, we propose the Belief-Calibrated Consensus Seeking (BCCS) framework to facilitate stable consensus via selecting optimal collaborators and calibrating the consensus judgment by system-internal beliefs. Experimental results on the MATH and MMLU benchmark datasets demonstrate that the proposed BCCS framework outperforms the best existing results by 2.23% and 3.95% of accuracy on challenging tasks, respectively. Our code and data are available at https://github.com/dengwentao99/BCCS.

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。

↑