让大模型智能体自动优化记忆，提升回复准确性和上下文理解。
MemInsight: Autonomous Memory Augmentation for LLM Agents
- 通过自动增强历史交互数据，改进记忆语义表示
- 在推荐任务中提升说服力14%，检索召回率比基线高34%
- 适合需要长期记忆与上下文理解的对话系统研究者
大型语言模型（LLM）智能体已具备信息处理、决策和与用户或工具交互的能力。其关键能力之一是整合长期记忆，以调用过往交互与知识。然而，记忆规模增长和语义结构化需求带来了显著挑战。本文提出一种自主记忆增强方法 MemInsight，以提升语义数据表征与检索机制。通过自动增强历史交互，LLM 智能体展现出更准确、更具上下文感的回应。我们在三个任务场景中验证了该方法的有效性：对话推荐、问答和事件摘要。在 LLM-REDIAL 数据集上，MemInsight 将推荐说服力最高提升 14%；在 LoCoMo 检索任务中，相比 RAG 基线，召回率提升 34%。实证结果表明，MemInsight 能有效提升多任务场景下 LLM 智能体的上下文表现。
原文摘要 · Abstract (English)
Large language model (LLM) agents have evolved to intelligently process information, make decisions, and interact with users or tools. A key capability is the integration of long-term memory capabilities, enabling these agents to draw upon historical interactions and knowledge. However, the growing memory size and need for semantic structuring pose significant challenges. In this work, we propose an autonomous memory augmentation approach, MemInsight, to enhance semantic data representation and retrieval mechanisms. By leveraging autonomous augmentation to historical interactions, LLM agents are shown to deliver more accurate and contextualized responses. We empirically validate the efficacy of our proposed approach in three task scenarios; conversational recommendation, question answering and event summarization. On the LLM-REDIAL dataset, MemInsight boosts persuasiveness of recommendations by up to 14%. Moreover, it outperforms a RAG baseline by 34% in recall for LoCoMo retrieval. Our empirical results show the potential of MemInsight to enhance the contextual performance of LLM agents across multiple tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。