AI 中文总结
该研究针对大语言模型智能体的长期记忆问题，提出MemSIF框架，通过结构化交互记忆与双轨事实记忆缓解两种错位模式，在两个数据集上的五种主干模型中均取得最高总准确率。
AI 中文摘要
长期记忆对于在长周期交互中运行的大语言模型（LLM）智能体至关重要。然而，现有记忆系统存在的若干持续局限可追溯至长期交互场景中的两种反复出现的错位模式：时间-结构错位（TSM）与延迟效用显现（DUM）。TSM指时间邻近性无法可靠匹配主题或事件层面相关性，DUM指写入时的显著性无法可靠预测未来查询效用。为缓解这些错位模式，我们提出MemSIF（结构化交互与事实记忆），即一种从结构化交互到事实记忆的框架。结构化交互记忆将原始交互组织为保留局部主题连贯性的主题片段，以及维持跨时间事件连续性的事件轨迹。双轨事实记忆采用两条互补轨道：核心事实（CoreFact）记忆在写入时整合稳定的、模式引导的信息，而活跃事实（ActiveFact）记忆按需形成事实，并将得到多个历史来源支持且受重复查询需求的事实提升以便复用。在LoCoMo和LongMemEval-S数据集上针对五种主干大语言模型开展的实验表明，MemSIF在所有设置中均取得最高的总准确率（Total ACC），在LoCoMo上较最强基线的提升为2.29%-8.79%，在LongMemEval-S上的提升为2.87%-6.15%。这些结果支持将结构化交互记忆与双轨事实记忆结合以缓解TSM和DUM的有效性，代码可访问指定URL获取。
英文摘要
Long-term memory is critical for LLM agents operating over long-horizon interactions. However, several persistent limitations of existing memory systems can be traced to two recurring misalignment patterns in long-term interaction settings: Temporal-Structural Misalignment (TSM) and Delayed Utility Manifestation (DUM). TSM arises when temporal proximity does not reliably align with topical or event-level relatedness, whereas DUM arises when write-time salience does not reliably predict future query utility. To mitigate these misalignment patterns, we propose MemSIF (Memory with Structured Interactions and Facts), a structured interaction-to-fact memory framework. Structured Interaction Memory organizes raw interactions into Topical Segments that preserve local topical coherence and Event Trajectories that maintain cross-time event continuity. Dual-Track Fact Memory uses two complementary tracks: CoreFact memory consolidates stable, schema-guided information at write time, whereas ActiveFact memory forms facts on demand and promotes those supported by multiple historical sources and recurring query demand for reuse. Experiments on LoCoMo and LongMemEval-S across five backbone LLMs show that MemSIF achieves the highest Total ACC in all settings, outperforming the strongest baseline by 2.29%-8.79% on LoCoMo and 2.87%-6.15% on LongMemEval-S. These results support the effectiveness of combining Structured Interaction Memory with Dual-Track Fact Memory to mitigate TSM and DUM. Code is available at https://github.com/luoyufeihaha/MemSIF.
CommentsSubmitted to AAAI 2027. 19 pages, 10 figures, 18 tables
