[论文] From Relevant Memories to Sufficient Evidence: Rethinking Memory Ut...

论文概要

研究领域: ML
作者: Shixuan Li, Wei Yang, Peiyu Zhang, Anzhe Cheng, Heng Ping, Xiaole Zhang, Jesse Thomason, Xuezhe Ma, Yue Zhao, Paul Bogdan
发布时间: 2026-10-19
arXiv: 2610.22668

中文摘要

长期记忆让 LLM 智能体得以复用过往经验，但检索到「相关」记忆未必提供回答查询的「充分」证据：关键信息可能散落在多条记录中、被抽象遮蔽、或与推理所需的关系脱节。本文指出这一「相关-充分缺口」，提出 QEMem 框架：把记忆利用重构为构建查询专属的证据集，从单条检索的相关性转向集体的证据支撑。三个互补机制：查询引导的发现——跨多种记忆粒度检索证据；源 grounded 的具体化——把证据组织成带显式证据角色与来源引用的结构化 ViewCard；充分性引导的精炼——选择性补回缺失信息。QEMem 直接运行在现有记忆库上，不改动记忆的形成方式。Claude Opus 4.5 下在 LoCoMo 上达到 91.43% 的 LLM 评审准确率，在共享记忆库下超过最强基线 MemR³ 7.6 个百分点、超 NEMORI 10.7 个百分点。进一步分析显示证据覆盖率改善，且在需要跨会话证据整合的问题上表现强劲。

原文摘要

Long-term memory enables LLM agents to reuse past experiences, yet retrieving relevant memories does not necessarily provide sufficient evidence for answering a query. Essential information may remain fragmented across memory records, obscured by abstraction, or disconnected from the relations required for reasoning. We identify this relevance-sufficiency gap and propose QEMem, a framework that reformulates memory utilization as the construction of query-specific evidence sets, shifting the focus from individual retrieval relevance to collective evidential support. QEMem integrates three complementary mechanisms: query-guided discovery retrieves evidence across multiple memory granularities; source-grounded reification organizes it into structured ViewCards with explicit evidential roles and source references; and sufficiency-guided refinement selectively recovers missing information. Operating on existing memory stores without modifying memory formation, QEMem achieves 91.43% LLM-judge accuracy on LoCoMo with Claude Opus 4.5, outperforming the strongest evaluated baseline, MemR\(^3\), by 7.6 percentage points and NEMORI by 10.7 points under a shared memory store. Further analyses demonstrate improved evidence coverage and strong performance on questions requiring cross-session evidence integration.

自动采集于 2026-10-19

#论文 #arXiv #ML #小凯