让大模型记忆更像人：用情景还原提升跨会话回忆能力
Drawing on Memory: Dual-Trace Encoding Improves Cross-Session Recall in LLM Agents
- 为每个事实搭配具体情境描述，增强记忆的上下文丰富性
- 跨会话回忆准确率提升至73.7%，较传统方法高20.2个百分点
- 特别适合需要追踪变化和多轮推理的智能体应用
具有持久记忆的大型语言模型通常将信息以扁平的事实记录方式存储，缺乏时间推理、变化追踪和跨会话聚合所需的上下文。受‘书写效应’启发，我们提出双痕迹记忆编码：每个存储的事实均配以具体的场景痕迹——即学习时刻的情境叙事重建。该方法强制模型在编码阶段记住具体细节，形成更丰富、更具区分度的记忆痕迹。基于LongMemEval-S基准（4,575个会话，100个召回问题），我们在99个共用问题上对比了双痕迹编码与仅事实控制组（覆盖范围与格式匹配）。双痕迹编码整体准确率达73.7%，较对照组53.5%提升20.2个百分点（95%置信区间：[+12.1, +29.3]，置换检验p < 0.0001）。增益主要集中在时间推理（+40pp）、知识更新追踪（+25pp）和多会话聚合（+30pp），单会话召回无提升，符合编码特异性理论。分词分析显示，该提升未增加额外计算开销。我们还初步设计了适用于编程智能体的双痕迹架构，并进行了验证。
原文摘要 · Abstract (English)
LLM agents with persistent memory store information as flat factual records, providing little context for temporal reasoning, change tracking, or cross-session aggregation. Inspired by the drawing effect [3], we introduce dual-trace memory encoding. In this method, each stored fact is paired with a concrete scene trace, a narrative reconstruction of the moment and context in which the information was learned. The agent is forced to commit to specific contextual details during encoding, creating richer, more distinctive memory traces. Using the LongMemEval-S benchmark (4,575 sessions, 100 recall questions), we compare dual-trace encoding against a fact-only control with matched coverage and format over 99 shared questions. Dual-trace achieves 73.7% overall accuracy versus 53.5%, a +20.2 percentage point (pp) gain (95% CI: [+12.1, +29.3], bootstrap p < 0.0001). Gains concentrate in temporal reasoning (+40pp), knowledge-update tracking (+25pp), and multi-session aggregation (+30pp), with no benefit for single-session retrieval, consistent with encoding specificity theory [8]. Token analysis shows dual-trace encoding achieves this gain at no additional cost. We additionally sketch an architectural design for adapting dual-trace encoding to coding agents, with preliminary pilot validation.
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。