缓解可靠高效搜索智能体的上下文干扰

中文摘要 AI 辅助

近期研究将大型语言模型（LLMs）赋能为多轮搜索智能体，通过迭代检索并生成输出直至解决复杂任务。然而，多轮搜索智能体的上下文冗长且复杂，例如每轮检索的文档集中难免引入无关信息，干扰LLMs，即所谓的“上下文干扰”，可能阻碍搜索智能体的可靠性与效率。因此，本文针对多轮搜索智能体的上下文干扰开展系统研究，重点探究三个问题：i）搜索智能体的上下文哪部分会引发上下文干扰；ii）如何优化搜索智能体的上下文以缓解干扰；iii）将上下文优化纳入搜索智能体的训练能否带来进一步提升。研究发现，干扰主要源于最新检索的文档。基于该发现，本文提出一种基于蒸馏的上下文优化器，用于动态缓解多轮搜索智能体的上下文干扰。最后，本文验证将上下文优化纳入搜索智能体的RL训练流程，可显著提升其可靠性与效率。该研究凸显了缓解搜索智能体上下文干扰的重要性，为AI智能体开创了“先优化上下文再生成”的新范式。

英文摘要

Recent research empowers Large Language Models (LLMs) as multi-turn search agents to iteratively retrieve and generate outputs until complex tasks are solved. However, the contexts of multi-turn search agents are lengthy and complex. For example, the retrieved set of documents in each turn would inevitably introduce irrelevant information that distracts LLMs, referring to \textit{context interference}, potentially hindering the reliability and efficiency of search agents. Therefore, we conduct a systematic study on context interference in multi-turn search agents, focusing on investigating i) which parts of the context of search agents will contribute to the context interference, ii) how to refine the contexts of search agents to mitigate the interference, and iii) can incorporating context refinement into search agent training yield further improvements. We reveal that interference primarily arises from the latest retrieved documents. Based on the explored findings, we then introduce a distill-based context refiner to dynamically mitigate context interference for multi-turn search agents. Finally, we validate that incorporating context refinement into RL training pipelines of search agents can significantly enhance both reliability and efficiency. This study highlights the importance of mitigating context interference of search agents, inspiring a novel paradigm of ``refine context and then generate'' for AI agents.

发表机构

• The Chinese University of Hong Kong（香港中文大学）
• University College London（伦敦大学学院）
• Zhejiang University（浙江大学）
• The University of Hong Kong（香港大学）
• The University of Edinburgh（爱丁堡大学）

机构由 AI 辅助整理，请以论文原文为准。