为边缘设备设计的类人长期记忆系统，提升对话真实性和记忆能力。
Mnemosyne: An Unsupervised, Human-Inspired Long-Term Memory Architecture for Edge-Based LLMs
- 采用图结构存储与动态过滤机制，模拟人类记忆的保留与遗忘。
- 在医疗对话中实现65.8%的人类评估真实度胜率，远超基线RAG的31.1%。
- 适合长期对话场景，尤其适用于资源受限的边缘智能应用。
长期记忆对自然、真实的对话至关重要。然而，当前大语言模型的记忆系统依赖暴力扩展上下文或静态检索流程，在边缘设备上表现不佳。我们提出Mnemosyne，一种无监督、受人类记忆启发的边缘型大模型长期记忆架构。该系统采用图结构存储、模块化内容与冗余过滤、记忆提交与修剪机制，以及基于时间衰减和刷新过程的概率性召回。Mnemosyne还引入一个浓缩的“核心摘要”，从记忆图的固定长度子集高效提取用户个性等长期信息，如医疗场景中的康复目标与护理态度。相较于现有检索增强方法，Mnemosyne专为纵向医疗助手设计，能有效应对重复但时间上独立的对话挑战。在纵向医疗对话实验中，其盲评真实度与长期记忆能力胜率为65.8%，显著高于基线RAG的31.1%。同时在LoCoMo基准测试中，于时间推理与单跳检索任务上达到当前最高分。平均综合得分54.6%，在所有方法中位居第二，优于常用Mem0与OpenAI基线。这表明，通过轻量级、可迁移的无监督记忆架构，可在边缘设备上实现更准确的事实回忆、更强的时间推理与更自然的用户交互。
原文摘要 · Abstract (English)
Long-term memory is essential for natural, realistic dialogue. However, current large language model (LLM) memory systems rely on either brute-force context expansion or static retrieval pipelines that fail on edge-constrained devices. We introduce Mnemosyne, an unsupervised, human-inspired long-term memory architecture designed for edge-based LLMs. Our approach uses graph-structured storage, modular substance and redundancy filters, memory committing and pruning mechanisms, and probabilistic recall with temporal decay and refresh processes modeled after human memory. Mnemosyne also introduces a concentrated "core summary" efficiently derived from a fixed-length subset of the memory graph to capture the user's personality and other domain-specific long-term details such as, using healthcare application as an example, post-recovery ambitions and attitude towards care. Unlike existing retrieval-augmented methods, Mnemosyne is designed for use in longitudinal healthcare assistants, where repetitive and semantically similar but temporally distinct conversations are limited by naive retrieval. In experiments with longitudinal healthcare dialogues, Mnemosyne demonstrates the highest win rate of 65.8% in blind human evaluations of realism and long-term memory capability compared to a baseline RAG win rate of 31.1%. Mnemosyne also achieves current highest LoCoMo benchmark scores in temporal reasoning and single-hop retrieval compared to other same-backboned techniques. Further, the average overall score of 54.6% was second highest across all methods, beating commonly used Mem0 and OpenAI baselines among others. This demonstrates that improved factual recall, enhanced temporal reasoning, and much more natural user-facing responses can be feasible with an edge-compatible and easily transferable unsupervised memory architecture.
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。