arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.20022cs.AI
收藏

记忆“何人”,知晓“如何”:面向持久助手的个人记忆与程序记忆基准测试

Memorizing Who, Knowing How: Benchmarking Personal and Procedural Memory for Persistent Assistants

Zhejiang University(浙江大学);Ant Group(蚂蚁集团)

Maosheng Qin, Qi Zhang, Mingxuan Xia, Jinglei Chen, Jiexiang Wang, Sheng Guo, Cheng Peng, Haobo Wang

AI总结:

本研究推出PraxisMemBench基准,发现现有方法难以协同维护个人与程序记忆,进而提出PraxisMem模型,其会话成功率达40.78%,显著优于Claude Code原生记忆。

摘要:

持久个人智能体的记忆系统必须超越孤立记忆,同时维护关于服务对象的个人记忆,以及关于如何行动的程序记忆。关键在于,在基于框架的个人助手中,程序或技能本质上是用户条件化的:个人上下文决定了哪些程序适用、如何实例化以及何时需要变更。现有基准通常要么侧重个人记忆,要么侧重程序记忆,而对二者的交互研究不足。我们推出PraxisMemBench,这是一个面向日常生活和工作轨迹的基于框架的持久个人助手基准,其中个人记忆与程序记忆共同演化,后续任务依赖于二者。实验表明,单独维护两种记忆类型是不够的,完成度从单一类型依赖下的84.6%降至相互依赖下的77.9%。更广泛而言,在现有方法中,随着交互历史增长,会话相关失败从早期到后期逐渐增加(59.8%升至72.9%),这揭示了持续维护挑战:若无法为后续检索保留其演化的内容和组织,仅写入正确的记忆是不够的。为解决此问题,我们提出PraxisMem,它利用特权联合状态监督,将维护记忆内容和组织的能力内化到记忆管理器中,采用位置感知GRPO和动态残差路由OPSD。PraxisMem的整体会话成功率达40.78%,相比之下,Claude Code的原生记忆仅为20.60%。

英文摘要:

Memory systems for persistent personal agents must move beyond isolated memories, maintaining both personal memory about whom they serve and procedural memory about how to act. Crucially, in harness-based personal assistance, procedures or skills are inherently user-conditioned: personal context determines which procedures apply, how they are instantiated, and when they should change. Existing benchmarks typically emphasize either personal or procedural memory, leaving their interaction underexplored. We introduce PraxisMemBench, a harness-based benchmark for persistent personal assistance across daily-life and work trajectories, where personal and procedural memory evolve together and later tasks depend on both. Experiments show that maintaining two memory types separately is insufficient, with completion dropping from single-type to interdependent dependencies (84.6% to 77.9%). More broadly, across existing methods, relation-related failures increase from early to late sessions as interaction history grows (59.8% to 72.9%), revealing a continual-maintenance challenge: writing correct memories is not enough if their evolving content and organization cannot be preserved for later retrieval. To address this, we propose PraxisMem, which uses privileged joint-state supervision to internalize the ability to maintain memory content and organization into the memory manager, with location-aware GRPO and dynamic residual-routed OPSD. PraxisMem reaches 40.78% overall Session Success Rate, compared with 20.60% for Claude Code's native memory. Our code is publicly available at https://github.com/QMMMS/PraxisMemBench.

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。

↑