arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.08601cs.CLcs.AIcs.LGcs.MA
收藏

Mnemosyne:一种面向边缘LLM的无监督、受人类启发的长期记忆架构

Mnemosyne: An Unsupervised, Human-Inspired Long-Term Memory Architecture for Edge-Based LLMs

Kaliber AI, Research Division(Kaliber AI 研究部门)

Aneesh Jonelagadda, Christina Hahn, Haoze Zheng, Salvatore Penachio

AI总结:

提出Mnemosyne,一种受人类记忆启发的无监督长期记忆架构,用于边缘LLM,通过图存储、概率回忆和核心摘要,在纵向医疗对话中实现65.8%的胜率,优于RAG基线。

摘要:

长期记忆对于自然、逼真的对话至关重要。然而,当前的大语言模型(LLM)记忆系统要么依赖暴力上下文扩展,要么依赖静态检索流水线,这些方法在边缘受限设备上无法有效运行。我们提出了Mnemosyne,一种专为基于边缘的LLM设计的无监督、受人类启发的长期记忆架构。我们的方法采用图结构存储、模块化物质与冗余过滤器、记忆提交与修剪机制,以及模拟人类记忆的带有时间衰减和刷新过程的概率性回忆。Mnemosyne还引入了一个集中的“核心摘要”,该摘要从记忆图的固定长度子集中高效推导,以捕捉用户的个性及其他特定领域的长期细节,例如,以医疗保健应用为例,包括康复后的抱负和对护理的态度。与现有的检索增强方法不同,Mnemosyne专为纵向医疗保健助手设计,在这些场景中,重复且语义相似但时间上不同的对话受限于朴素检索。在纵向医疗保健对话的实验中,与基线RAG的31.1%胜率相比,Mnemosyne在盲人评估中实现了65.8%的最高真实感和长期记忆能力胜率。与其他同骨干技术相比,Mnemosyne在时间推理和单跳检索方面也达到了当前最高的LoCoMo基准分数。此外,54.6%的平均总体分数在所有方法中位居第二,超过了常用的Mem0和OpenAI基线等。这表明,通过一种边缘兼容且易于迁移的无监督记忆架构,可以实现改进的事实回忆、增强的时间推理以及更自然的面向用户的响应。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑