主动提取记忆，让大模型智能区分细节、事件与关系
Beyond Static Summarization: Proactive Memory Extraction for LLM Agents
- 分类型主动提取：按细节/事件/关系分开处理，避免信息混杂
- 检查完整性和原子级验证，减少记忆遗漏与幻觉
- 适合长期交互的个性化大模型应用，提升记忆可用性
在长期且个性化的交互中，记忆管理对大模型代理至关重要。现有方法多关注记忆的检索与使用，却忽视记忆如何提取。我们发现现有方法存在两大局限：一是提取过于‘提前’，代理在不知未来任务时就保存信息，单一摘要提示常将细节、事件与关系混在一起，导致有用信息丢失；二是提取通常为一次性操作，缺乏验证，错误和幻觉可能长期存在于记忆中。为此，我们提出 ProMem 框架，通过分离细节、事件与关系，并为每类采用不同提取策略，同时检查完整性以恢复遗漏事件，并在原子层面验证事实以减少幻觉。实验表明，ProMem 提升了记忆完整性和问答准确率，同时在质量与令牌成本间保持良好平衡。
原文摘要 · Abstract (English)
Memory management is vital for LLM agents in long-term and personalized interactions. Most previous work studies how to retrieve and use memory, but pays less attention to how memory is extracted. We find two main limitations in existing methods. First, extraction is "ahead-of-time": the agent saves information before it knows future tasks. A single summary prompt often mixes details, events, and relations, so useful information is lost. Second, extraction is usually one-off. Without verification, errors and hallucinations may stay in memory for a long time. To address these limitations, we propose ProMem, a proactive memory extraction framework. It separates details, events, and relations, and uses different extraction strategies for each type. It also checks completeness to recover missed events and verifies facts at the atomic level to reduce hallucinations. Experiments show that ProMem improves memory completeness and QA accuracy, while keeping a good balance between quality and token cost.
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。