让大模型长时对话更连贯，通过结构化记忆捕捉事件关联。
StructMem: Structured Memory for Long-Horizon Behavior in LLMs
- 分层结构记忆，同时保留事件细节并建立跨事件联系。
- 在LoCoMo数据集上提升多跳问答能力，减少70%以上token消耗。
- 适合需要长期推理的对话系统、智能助手研发者使用。
长时对话智能体需要能捕捉事件间关系的记忆系统，而非仅存储孤立事实，以支持时间推理和多跳问答。现有方法存在根本权衡：扁平记忆高效但无法建模关系结构，图结构记忆虽可实现结构化推理，但构建成本高且脆弱。为此，我们提出结构增强型分层记忆框架 StructMem，保持事件级绑定并诱导跨事件连接。通过时间锚定双重视角并周期性进行语义整合，StructMem 在 exttt{LoCoMo} 数据集上提升了时间推理与多跳问答性能，同时显著降低 token 使用量、API 调用次数和运行时间，相较以往记忆系统有明显优化。代码开源地址：https://github.com/zjunlp/LightMem。
原文摘要 · Abstract (English)
Long-term conversational agents need memory systems that capture relationships between events, not merely isolated facts, to support temporal reasoning and multi-hop question answering. Current approaches face a fundamental trade-off: flat memory is efficient but fails to model relational structure, while graph-based memory enables structured reasoning at the cost of expensive and fragile construction. To address these issues, we propose \textbf{StructMem}, a structure-enriched hierarchical memory framework that preserves event-level bindings and induces cross-event connections. By temporally anchoring dual perspectives and performing periodic semantic consolidation, StructMem improves temporal reasoning and multi-hop performance on \texttt{LoCoMo}, while substantially reducing token usage, API calls, and runtime compared to prior memory systems, see https://github.com/zjunlp/LightMem .
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。