让大模型主动重构记忆，比死记硬背更懂上下文。
Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents
- 用图结构存储记忆，通过语义标签连接线索和内容
- 推理中动态调整记忆访问路径，准确率最高提升23%
- 适合长对话、复杂推理场景，节省计算资源
尽管近期取得进展，大模型智能体在处理长交互历史时仍存在困难。现有记忆增强型智能体依赖静态的‘检索-推理’流程，难以根据推理过程中发现的中间证据动态调整记忆访问。为此，我们提出MRAgent框架，结合关联记忆图与主动重构机制。将记忆表示为包含线索-标签-内容的图结构，其中语义标签作为连接细粒度线索与记忆内容的桥梁。在此结构上，主动重构机制将大模型推理直接融入记忆访问过程，使智能体能基于累积证据迭代探索并剪枝检索路径，确保记忆检索动态适配推理上下文，同时避免因无约束扩展导致的组合爆炸。在LoCoMo与LongMemEval两个基准上的实验表明，相比强基线模型，该方法在准确率上最高提升23%，同时显著降低令牌消耗与运行时间，验证了主动且关联式重构在长周期记忆推理中的有效性。
原文摘要 · Abstract (English)
Despite recent progress, LLM agents still struggle with reasoning over long interaction histories. While current memory-augmented agents rely on a static retrieve-then-reason paradigm, this rigid pipeline design prevents them from dynamically adapting memory access to intermediate evidence discovered during inference. To bridge this gap, we propose MRAgent, a framework that combines an associative memory graph with an active reconstruction mechanism. We represent memory as a Cue-Tag-Content graph, where associative tags serve as semantic bridges connecting fine-grained cues to memory contents. Operating on this structure, our active reconstruction mechanism integrates LLM reasoning directly into memory access, allowing the agent to iteratively explore and prune retrieval paths based on accumulated evidence. This ensures that memory retrieval is dynamically adapted to the reasoning context while avoiding combinatorial explosion caused by unconstrained expansion. Experiments on the LoCoMo benchmark and LongMemEval benchmark demonstrate significant improvements over strong baselines (up to 23%), while substantially reducing token and runtime cost, highlighting the effectiveness of active and associative reconstruction for long-horizon memory reasoning.
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。