arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.07582cs.CLcs.AI
收藏

基于事件分割的记忆:长期对话代理中的记忆

ES-Mem: Event Segmentation-Based Memory for Long-Term Dialogue Agents

College of Computer Science, Chongqing University(重庆大学计算机科学学院);Tsinghua University(清华大学);Hong Kong Generative AI Research & Development Center, HKUST(香港科技大学生成式人工智能研究与开发中心);Aerospace Information Research Institute, Chinese Academy of Sciences(中国科学院航空航天信息研究所);School of Computer Science and Technology, Tiangong University(天津理工大学计算机科学与技术学院)

Huhai Zou, Tianhao Sun, Chuanjiang He, Yu Tian, Zhenyang Li, Li Jin, Nayu Liu, Jiang Zhong, Kaiwen Wei

AI总结:

ES-Mem通过动态事件分割和分层记忆架构,提升长期对话代理的记忆效率和上下文定位能力。

摘要:

记忆对于对话代理在长期交互中保持连贯性和实现连续适应至关重要。尽管现有的记忆机制提供了基本的存储和检索能力,但它们受到两个主要限制:(1)刚性的记忆粒度常常破坏语义完整性,导致碎片化和不连贯的记忆单元;(2)普遍的平面检索范式仅依赖于表层的语义相似性,忽略了对话中所需的结构线索,以导航和定位特定的事件背景。为了缓解这些限制,受事件分割理论的启发,我们提出了ES-Mem,一个包含两个核心组件的框架:(1)一个动态事件分割模块,将长期交互分割成具有明确边界的语义连贯事件;(2)一个分层记忆架构,构建多层次的记忆并利用边界语义来锚定特定的事件记忆以实现精确的上下文定位。在两个记忆基准测试中,ES-Mem在基准方法上显示出一致的性能提升。此外,所提出的事件分割模块在对话分割数据集上表现出稳健的应用性。

英文摘要:

Memory is critical for dialogue agents to maintain coherence and enable continuous adaptation in long-term interactions. While existing memory mechanisms offer basic storage and retrieval capabilities, they are hindered by two primary limitations: (1) rigid memory granularity often disrupts semantic integrity, resulting in fragmented and incoherent memory units; (2) prevalent flat retrieval paradigms rely solely on surface-level semantic similarity, neglecting the structural cues of discourse required to navigate and locate specific episodic contexts. To mitigate these limitations, drawing inspiration from Event Segmentation Theory, we propose ES-Mem, a framework incorporating two core components: (1) a dynamic event segmentation module that partitions long-term interactions into semantically coherent events with distinct boundaries; (2) a hierarchical memory architecture that constructs multi-layered memories and leverages boundary semantics to anchor specific episodic memory for precise context localization. Evaluations on two memory benchmarks demonstrate that ES-Mem yields consistent performance gains over baseline methods. Furthermore, the proposed event segmentation module exhibits robust applicability on dialogue segmentation datasets.

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。

↑