arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.09893cs.CL
收藏

面向知识密集型任务的具有轨迹学习的协同多智能体框架

Synergistic Multi-Agent Framework with Trajectory Learning for Knowledge-Intensive Tasks

Fudan University(复旦大学);University of Southern California(南加州大学);Huazhong University of Science and Technology(华中科技大学)

Shengbin Yue, Siyuan Wang, Wei Chen, Xuanjing Huang, Zhongyu Wei

AI总结:

本文提出多智能体框架SMART,通过四个专门智能体执行子轨迹动作并采用长短轨迹学习进行协同训练,利用外部知识提升知识密集型任务中LLM响应的事实一致性与可解释性,实验证明其优于现有方法。

摘要:

大语言模型(LLMs)的最新进展在各种自然语言处理任务中带来了重大突破。然而,在知识密集型场景中生成事实一致的响应仍然是一个挑战,原因包括幻觉问题、难以获取长尾知识以及记忆扩展有限。本文介绍了SMART,这是一个新颖的多智能体框架,利用外部知识来增强LLM生成响应的可解释性和事实一致性。SMART包含四个专门的智能体,每个智能体执行特定的子轨迹动作,以应对复杂的知识密集型任务。我们提出了一种多智能体协同训练范式,即长短轨迹学习,它确保智能体之间的协同合作,同时保持每个智能体的细粒度执行。在五个知识密集型任务上的大量实验表明,与广泛采用的知识内化和知识增强方法相比,SMART具有更优越的性能。我们的框架可以扩展到知识密集型任务之外的更复杂场景。我们的代码可在https://github.com/yueshengbin/SMART获取。

英文摘要:

Recent advancements in Large Language Models (LLMs) have led to significant breakthroughs in various natural language processing tasks. However, generating factually consistent responses in knowledge-intensive scenarios remains a challenge due to issues such as hallucination, difficulty in acquiring long-tailed knowledge, and limited memory expansion. This paper introduces SMART, a novel multi-agent framework that leverages external knowledge to enhance the interpretability and factual consistency of LLM-generated responses. SMART comprises four specialized agents, each performing a specific sub-trajectory action to navigate complex knowledge-intensive tasks. We propose a multi-agent co-training paradigm, Long-Short Trajectory Learning, which ensures synergistic collaboration among agents while maintaining fine-grained execution by each agent. Extensive experiments on five knowledge-intensive tasks demonstrate SMART's superior performance compared to widely adopted knowledge internalization and knowledge enhancement methods. Our framework can extend beyond knowledge-intensive tasks to more complex scenarios. Our code is available at https://github.com/yueshengbin/SMART.

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。

↑