对比图结构与非图结构记忆系统，发现基础设置比架构创新更重要
Does Memory Need Graphs? A Unified Framework and Empirical Analysis for Long-Term Dialog Memory
- 构建统一框架，拆解对话记忆为可独立测试的组件
- 在两个数据集上验证，多数性能差异源于系统配置而非图结构设计
- 提出稳定可靠的基线方法，助力未来对话记忆研究
图结构在对话记忆系统中越来越常见，但其实际效果仍存争议。本文从系统角度开展实验分析，提出一个统一框架，将对话记忆系统分解为核心模块，支持图结构与非图结构方法的对比。基于该框架，在LongMemEval和HaluMem数据集上进行分阶段控制实验，对比记忆表示、组织、维护与检索等常见设计选择。结果表明，多数性能差异由基础系统设置驱动，而非特定架构创新。据此，我们识别出稳定可靠的强基线方法，为未来对话记忆研究提供参考。代码已开源。
原文摘要 · Abstract (English)
Graph structures are increasingly used in dialog memory systems, but empirical findings on their effectiveness remain inconsistent, making it unclear which design choices truly matter. We present an experimental, system-oriented analysis of long-term dialog memory architectures. We introduce a unified framework that decomposes dialog memory systems into core components and supports both graph-based and non-graph approaches. Under this framework, we conduct controlled, stage-wise experiments on LongMemEval and HaluMem, comparing common design choices in memory representation, organization, maintenance, and retrieval. Our results show that many performance differences are driven by foundational system settings rather than specific architectural innovations. Based on these findings, we identify stable and reliable strong baselines for future dialog memory research. Code are available at https://github.com/AvatarMemory/UnifiedMem
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。