arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.04780cs.CL
收藏

PersonaTree: 面向LLM智能体人物理解的结构化生命周期记忆

PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents

School of ASEE, Beihang University, Beijing, China(北京航空航天大学航空科学与工程学院);The University of Hong Kong, Hong Kong, China(香港大学);Peking University, Beijing, China(北京大学);University of Chinese Academy of Sciences, Beijing, China(中国科学院大学);School of BME, Beihang University, Beijing, China(北京航空航天大学生物医学工程学院);CAIR and CECS, VinUniversity, Hanoi, Vietnam(越南河内 Vin 大学 CAIR 和 CECS)

Yubo Hou, Jingwei Song, Hongbo Zhang, Zhisheng Chen, Bang Xiao, Tao Wan, Zengchang Qin

AI总结:

提出PersonaTree,一种结构化生命周期记忆框架,通过三级人物树和显式支持路径,将交互证据抽象为人物理解,在多个基准上取得领先性能。

摘要:

持久化的LLM智能体需要记忆表示,使得在长期交互中人物理解的形成变得明确。现有的智能体记忆方法强调信息保留和检索,但对累积的交互证据如何被抽象为人物理解的解释有限。我们将这一过程视为图式形成,其中情境证据被抽象为可重用模式和稳定的人物层面断言。我们引入PersonaTree,一种结构化生命周期记忆框架,通过三级人物树实现这一观点,并具有从证据到断言的显式支持路径。PersonaTree通过保守写入、置信度引导的合并和查询条件路径检索来维护树,仅返回每个查询所需的证据深度。在六个涉及人物理解和持久记忆的基准测试中,使用三个回答骨干,PersonaTree在18个紧凑分数中排名第一,并在16个设置中进入前两名。消融实验表明,层次结构提高了KnowMe上的抽象人物理解,而在可比上下文预算下,支持路径检索提高了RealPref的对齐度。

英文摘要:

Persistent LLM agents require memory representations that make the formation of person understanding explicit across long term interaction. Existing agent memory methods emphasize information retention and retrieval, yet give limited account of how accumulated interaction evidence is abstracted into person understanding. We view this process as schema formation, where situated evidence is abstracted into reusable patterns and stable person level claims. We introduce PersonaTree, a structured lifecycle memory framework that realizes this view as a three level persona tree with explicit support paths from evidence to claims. PersonaTree maintains the tree through conservative writing, confidence guided consolidation, and query conditioned path retrieval, returning only the evidence depth required by each query. Across six person understanding and persistent memory benchmarks with three answer backbones, PersonaTree ranks first in 12 of 18 compact scores and reaches the top two in 16 settings. Ablations show that hierarchy improves abstract person understanding on KnowMe, while support path retrieval improves RealPref alignment under a comparable context budget.

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。

↑