ShadowMem：通过影子记忆保护 LLM 智能体免受长时程威胁

Computer Science > Cryptography and Security arXiv:2605.03228 (cs) [Submitted on 4 May 2026 (v1), last revised 23 Sep 2026 (this version, v2)] Title:Safeguarding LLM Agents against Long-Horizon Threats via Shadow Memory Authors:Yuhui Wang, Tanqiu Jiang, Jiacheng Liang, Charles Fleming, Ting Wang View a PDF of the paper titled Safeguarding LLM Agents against Long-Horizon Threats via Shadow Memory, by Yuhui Wang and 4 other authors View PDF HTML (experimental) Abstract:As large language model (LLM)-powered agents are increasingly deployed to perform complex, real-world tasks, they face a growing class of attacks that exploit extended user-agent-environment interactions to pursue malicious objectives improbable in single-turn settings. Such long-horizon threats pose significant risks to the safe deployment of LLM agents in critical domains. In this paper, we present ShadowMem, a novel defensive framework designed to counter a wide range of long-horizon threats. Inspired by the "shadow stack" abstraction in systems security, ShadowMem maintains a dedicated, safety-focused agentic memory that distills and retains safety-critical context across the agent's full execution trajectory, leveraging this shadow memory to proactively assess the risk of pending actions prior to their execution. Extensive evaluation demonstrates that ShadowMem substantially outperforms existing defenses across diverse long-horizon threats in detection accuracy, achieves early-stage detection for the majority of attacks, and introduces only negligible overhead to agent utility. To our best knowledge, ShadowMem represents the first framework to detect and mitigate long-horizon threats using an agentic memory approach, establishing a new paradigm for this critical challenge and opening promising directions for future research. The artifacts are available at this https URL Comments: Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2605.03228 [cs.CR] (or arXiv:2605.03228v2 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2605.03228 Focus to learn more arXiv-issued DOI via DataCite Related DOI: https://doi.org/10.1145/3830454.3846601 Focus to learn more DOI(s) linking to related resources Submission history From: Yuhui Wang [view email] [v1] Mon, 4 May 2026 23:46:19 UTC (2,140 KB) [v2] Wed, 23 Sep 2026 17:47:23 UTC (2,035 KB) Full-text links: Access Paper: View a PDF of the paper titled Safeguarding LLM Agents against Long-Horizon Threats via Shadow Memory, by Yuhui Wang and 4 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CR new | recent | 2026-05 Change to browse by: cs cs.AI cs.CL References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)