arXiv cs.CL· Haoming Li, Dingyi Kang, Xu Hu, Jinxin Yang, Zhu Xu, Chenru Wang, Yi Ge, Miao Yin, Guanpeng Li, Feng Chen, Bingzhe Li·· 12 小时前AI 评分39
SAFE-Mem：带细粒度权限与治理感知检索的安全智能体记忆框架
SAFE-Mem: Secure Agentic Memory with Fine-Grained Authority and Governance-Aware Retrieval
AI 导读
SAFE-Mem 是一个细粒度记忆治理框架，为每条声明标注 kind、authority、scope 三个治理标签，并通过治理感知检索将查询需求与声明权限匹配，阻止未授权影响。在 MPBench 和 LoCoMo 上、四种 LLM 基座下，其攻击成功率比保留合法记忆的最强防御方法至少低 77%，答案准确率最高提升 15%。代码已开源。
正文 · 原文
Authors:Haoming Li, Dingyi Kang, Xu Hu, Jinxin Yang, Zhu Xu, Chenru Wang, Yi Ge, Miao Yin, Guanpeng Li, Feng Chen, Bingzhe Li
Abstract:Large language model (LLM) agents increasingly rely on persistent memory to support long-term reasoning and personalized interactions. However, information collected from external sources can introduce security risks when untrusted claims (i.e., atomic statements) are later treated as authorized preferences, policies, or permissions. Existing defenses typically rely on content-based filtering or coarse-grained provenance tracking, which either fail to prevent unauthorized memory use or block legitimate information, sacrificing memory utility. To address these limitations, we propose SAFE-Mem, a fine-grained memory governance framework that improves security while preserving legitimate memory use. First, SAFE-Mem annotates each claim with three governance labels (i.e., kind, authority, and scope) to capture its provenance and permitted use. Second, governance-aware retrieval matches query requirements with claim permissions to block unauthorized influence while retaining legitimate information. Third, a governance-enabled memory graph connects source sentences, claims, and entities to support provenance tracking and efficient retrieval. Across four LLM backbones on MPBench and LoCoMo, SAFE-Mem attains the lowest attack success among defenses that retain legitimate memory, at least 77% below the strongest of them, and the highest answer accuracy of all evaluated defenses, improving it by up to 15%. The code is available at this https URL.
| Subjects: | Cryptography and Security (cs.CR); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.22959 [cs.CR] |
| (or arXiv:2610.22959v1 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22959 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Haoming Li [view email]
[v1]
Fri, 16 Oct 2026 05:29:55 UTC (328 KB)
来源：arXiv cs.CL · arxiv.org