让大模型记忆系统自己进化，自动优化检索策略。
EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents
- 用大模型诊断问题，自动调整记忆检索配置。
- 在LoCoMo上比最强基线提升25.7%，在MemBench上提升18.9%。
- 自进化出通用检索机制，跨数据集迁移效果好。
长期记忆对跨会话运行的大语言模型代理至关重要，但现有记忆系统将检索基础设施视为固定：存储内容可变，而评分函数、融合策略和答案生成策略在部署后保持不变。我们提出，真正自适应的记忆需要在两个层面协同演化：存储知识与查询它的检索机制。EvolveMem是一种自演化记忆架构，将完整检索配置作为结构化动作空间，由大模型驱动的诊断模块进行优化。每轮演化中，该模块读取每个问题的失败日志，识别根本原因，并提出针对性配置调整；一个受保护的元分析器在自动回滚异常和停滞时探索新方向的保障下应用这些调整。这一闭环自演化实现了AutoResearch过程：系统自主对其自身架构进行迭代研究，取代人工配置调优。从最小基线开始，该过程自主收敛，发现了包括原始动作空间中不存在的新配置维度在内的有效检索策略。在LoCoMo上，EvolveMem相比最强基线相对提升25.7%，比最小基线提升78.0%。在MemBench上，超越最强基线18.9%。演化的配置在不同基准间实现正向迁移而非灾难性迁移，表明自演化过程捕捉的是通用检索原理而非特定基准的技巧。代码见https://github.com/aiming-lab/SimpleMem。
原文摘要 · Abstract (English)
Long-term memory is essential for LLM agents that operate across multiple sessions, yet existing memory systems treat retrieval infrastructure as fixed: stored content evolves while scoring functions, fusion strategies, and answer-generation policies remain frozen at deployment. We argue that truly adaptive memory requires co-evolution at two levels: the stored knowledge and the retrieval mechanism that queries it. We present EvolveMem, a self-evolving memory architecture that exposes its full retrieval configuration as a structured action space optimized by an LLM-powered diagnosis module. In each evolution round, the module reads per-question failure logs, identifies root causes, and proposes targeted configuration adjustments; a guarded meta-analyzer applies them with automatic revert-on-regression and explore-on-stagnation safeguards. This closed-loop self-evolution realizes an AutoResearch process: the system autonomously conducts iterative research cycles on its own architecture, replacing manual configuration tuning. Starting from a minimal baseline, the process converges autonomously, discovering effective retrieval strategies including entirely new configuration dimensions not present in the original action space. On LoCoMo, EvolveMem outperforms the strongest baseline by 25.7% relative and achieves a 78.0% relative improvement over the minimal baseline. On MemBench, EvolveMem exceeds the strongest baseline by 18.9% relative. Evolved configurations transfer across benchmarks with positive rather than catastrophic transfer, indicating that the self-evolution process captures universal retrieval principles rather than benchmark-specific heuristics. Code is available at https://github.com/aiming-lab/SimpleMem.
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。