Thought-Retriever: 不只是检索原始数据，为记忆增强型智能体检索思想
Thought-Retriever: Don't Just Retrieve Raw Data, Retrieve Thoughts for Memory-Augmented Agentic Systems
AI总结：
本文提出Thought-Retriever，一种无需受上下文长度限制的算法，使LLM能基于任意长外部数据生成输出。通过利用历史查询生成的中间响应，构建长期记忆，提升回答能力。
摘要：
大型语言模型（LLMs）因强大的内部能力和知识而改变了AI研究，但现有LLMs在与世界交互时仍无法有效整合大量外部知识。尽管提出了检索增强型LLMs，但其仍受限于LLM的上下文长度，只能检索外部知识库中的前K个原始数据块。本文提出Thought-Retriever，一种模型无关的算法，帮助LLM生成输出时不受上下文长度或检索数据块数量的限制。我们的关键洞察是让LLM充分利用解决过去用户查询时生成的中间响应（思想），过滤无意义和冗余的思想，将其组织在思想记忆中，并在处理新查询时检索相关思想。这有效地为基于LLM的智能体配备了自我进化长期记忆，通过持续交互不断增长其能力。此外，我们进一步精心准备了一个新的基准测试集AcademicEval，要求LLM基于真实世界学术论文利用超长上下文回答查询。在AcademicEval和两个其他公开数据集上的广泛实验验证了Thought-Retriever在各种任务中显著优于现有最先进基线，平均F1分数增加至少7.6%，胜率增加16%。更重要的是，我们进一步展示了两个激动人心的发现：（1）Thought-Retriever确实在解决更多用户查询后能够帮助LLM自我进化；（2）Thought-Retriever学会利用更深入的思想来回答更抽象的用户查询。
英文摘要：
Large language models (LLMs) have transformed AI research thanks to their powerful internal capabilities and knowledge. However, existing LLMs still fail to effectively incorporate the massive external knowledge when interacting with the world. Although retrieval-augmented LLMs are proposed to mitigate the issue, they are still fundamentally constrained by the context length of LLMs, as they can only retrieve top-K raw data chunks from the external knowledge base which often consists of millions of data chunks. Here we propose Thought-Retriever, a novel model-agnostic algorithm that helps LLMs generate output conditioned on arbitrarily long external data, without being constrained by the context length or number of retrieved data chunks. Our key insight is to let an LLM fully leverage its intermediate responses generated when solving past user queries (thoughts), filtering meaningless and redundant thoughts, organizing them in thought memory, and retrieving the relevant thoughts when addressing new queries. This effectively equips LLM-based agents with a self-evolving long-term memory that grows more capable through continuous interaction. Besides algorithmic innovation, we further meticulously prepare a novel benchmark, AcademicEval, which requires an LLM to faithfully leverage ultra-long context to answer queries based on real-world academic papers. Extensive experiments on AcademicEval and two other public datasets validate that Thought-Retriever remarkably outperforms state-of-the-art baselines, achieving an average increase of at least 7.6% in F1 score and 16% in win rate across various tasks. More importantly, we further demonstrate two exciting findings: (1) Thought-Retriever can indeed help LLM self-evolve after solving more user queries; (2) Thought-Retriever learns to leverage deeper thoughts to answer more abstract user queries.
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。
