Skip to content

Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

All papers

Found 2 papers
Filter3

Defense type

Detection and filtering2

System component

Memory2Guardrails and judges32Monitors3Vision4Video3Embeddings0Multimodal8
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 233papers have no tags yet and are not included in filtered results.
Defense type：Detection and filteringSystem component：MemorySystem component：EmbeddingsClear all

Another 233papers do not have category tags yet and are not included in filtered results.

  1. Evaluation and benchmarksarXiv 2609.08258

    A study empirically tests five Agent memory systems and finds none of them enforce revocation markers

    评测 Agent 记忆系统的软撤回执行情况，并提出陈旧检索防护

    arXiv
    2609.08258
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5.5, Grok-4, DeepSeek-V4-Flash and others9

    Abstract系统揭示了记忆系统软撤回在检索时普遍未执行的安全漏洞，证明写回会击穿现有防御，并提出读取端防护作为必要控制手段。

    Paper analysisRead analysis
  2. Evaluation and benchmarksarXiv 2604.01350

    No Attacker Needed: Unintentional Cross-User Contamination in Shared-State LLM Agents

    形式化并评测共享状态 Agent 的无意跨用户污染

    arXiv
    2604.01350
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-4o
    Abstract作者报告良性交互即可造成跨用户污染，文本清洗挡不住代码记忆。

    UCC 是共享状态 LLM 智能体上的稳健性问题：源交互可以是良性的，局部有效的制品仍会在后来被误用到其他用户。

    Paper analysisRead analysis
You have reached the end