Skip to content

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

Oct 12 · Monday

All papers

Found 4 papers
Filter4

System type

AI Agent3

Attack surface / entry point

User input22Web pages and interfaces5Documents and messages10Tool responses6Tool and skill descriptions3Code repositories and dependencies4Retrieval corpus4Memory writes16Images, audio and video4

Layer

Application layer3

System component

Memory2Guardrails and judges3Monitors0Reasoning chains2
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 240papers have no tags yet and are not included in filtered results.
Attack surface / entry point：Code repositories and dependenciesSystem component：MemorySystem component：MonitorsSystem component：Reasoning chainsClear all

Another 240papers do not have category tags yet and are not included in filtered results.

  2. AttacksarXiv 2608.09867

    Researchers steal proprietary LLM inference chains via cross-model compatibility of encrypted inference blocks

    利用加密思维块的跨模型重放，诱导弱模型逐字转写专有推理链

    arXiv
    2608.09867
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Claude Opus 4.8, Claude Sonnet 4.6, Claude Haiku 4.5 and others13

    Abstract论文揭示了客户端加密思维块跨模型互换漏洞，提出了低成本提取方法并给出防御方案。

    Paper analysisRead analysis
  4. AttacksarXiv 2605.15338

    Hidden in Memory: Latent Memory Poisoning of LLM Agents

    提出潜伏记忆投毒，诱导 Agent 把外部文档写成持久伪造记忆

    arXiv
    2605.15338
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5.4, GPT-5.5, Claude Sonnet 4.6 and others12
    Abstract持久记忆成为跨会话攻击面：通用载荷可写入伪造记忆，并在相关后续任务中影响行为。

    sleeper memory poisoning 研究带持久记忆的 LLM 助手：外部文档能否让助手存下一则伪造的用户记忆，并在原文档消失后的新会话里被取回、用来改变回复或动作。

    Paper analysisRead analysis
You have reached the end