Skip to content

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

Oct 12 · Monday

All papers

Found 12 papers
Filter5

Attack type

Jailbreaks12Prompt injection92Poisoning and backdoors50Extraction and theft3Detection evasion3Malicious use6

Defense type

Instruction–data separation3Permissions and approval5Monitoring and auditing4

Impact

Harmful content3Unauthorized actions6Data and privacy leaks3

System type

AI Agent12

Attack surface / entry point

User input12Tool responses5

Layer

Model layer7Application layer12

System component

Memory3Guardrails and judges10Monitors4Permissions and sandboxing5
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 254papers have no tags yet and are not included in filtered results.
Attack type：JailbreaksLayer：Application layerSystem component：MemorySystem component：MonitorsSystem component：Permissions and sandboxingClear all

Another 254papers do not have category tags yet and are not included in filtered results.

  1. Evaluation and benchmarksarXiv 2610.07089

    研究者提出统一滥用监控基准，覆盖分解攻击与提示注入约 6200 条轨迹

    提出危害窗口与区间指标，统一评测分解攻击和提示注入下的智能体滥用监视

    arXiv
    2610.07089
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Gemma 4 26B-A4B, Qwen3.6 27B, Qwen3.5 27B and others12

    Abstract统一智能体滥用监视框架，动作视角兼顾两类威胁但危害定位仍存瓶颈。

    Paper analysisRead analysis
  4. DefensesarXiv 2609.34518

    SEAD 论文提出工具型 Agent 的状态视角攻击与防御方法 DART 与 SAGE

    提出工具型 Agent 的自适应攻击 DART 与预执行防御 SAGE

    arXiv
    2609.34518
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5.6 Luna, GPT-5.6 Terra, Gemini 3.8 Flash and others5

    Abstract论文提出状态控制框架 SEAD、攻击 DART 与防御 SAGE，实验显示执行反馈与环境探测在攻防交互中具有关键作用。

    Paper analysisRead analysis
  5. DefensesarXiv 2609.28693

    skilder：以角色化技能包为工具型 LLM 智能体做访问控制

    提出 Skilder，按角色渐进下发企业工具并确定性拦截越权调用

    arXiv
    2609.28693
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Claude Haiku 4.5, Qwen 3.5 122B, Gemma 4 31B and others6
    AbstractSkilder 用单 MCP 服务器按角色交付工具并硬拦截。

    Skilder 把企业工具型智能体的能力按角色打包，经一个 MCP 服务器做渐进发现，使工具范围由服务器而不是提示词来执行。

    Paper analysisRead analysis
  7. DefensesarXiv 2609.14003

    FlowSeal：用信息流控制阻断 LLM Agent 的隐私泄露

    提出 FLOWSEAL，用工具级信息流控制阻断智能体隐私泄露

    arXiv
    2609.14003
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    DeepSeek-V3.2, GPT-4.1-mini, Gemini-Flash-2.5

    Abstract作者分析大模型智能体隐私检查与攻击面重合问题，提出外部拦截防御 FLOWSEAL 并验证有效性。

    Paper analysisRead analysis
  8. OtherarXiv 2609.12413

    SoK 综述：智能体 AI 时代的越狱攻击、防御与实践考量

    系统梳理智能体执行链路中的越狱攻击与防御并比较推理时防御

    arXiv
    2609.12413
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Qwen3.5-4B, Llama-3.1-8B-Instruct
    AbstractSoK 将越狱重放到智能体流水线，并用两种模型检验回复层防御的边际与代价。

    Mia 等人把越狱安全从单轮回复扩展到智能体的规划、记忆、工具和智能体间通信，并问既有结论在更强原生对齐下是否还成立。

    Paper analysisRead analysis
  10. AttacksarXiv 2606.09084

    研究者提出 Context-Fractured Decomposition 攻击

    提出 CFD 跨会话分解攻击，利用工件来源缺口绕过工具型 Agent 的本地检查

    arXiv
    2606.09084
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-4.1, GPT-5-nano, Claude-3.5-Sonnet and others6
    AbstractCFD 用无指令 artifact 把越狱拆到跨会话。

    CFD针对工具型 LLM 智能体：利用 artifact 来源不被跟踪这一部署失败模式，把有害目标拆开后延迟组合。

    Paper analysisRead analysis
  12. DefensesarXiv 2505.23559

    SafeScientist：让 LLM 智能体的科学发现具备风险意识

    提出 SafeScientist 四层监控，阻止科研智能体执行高风险或不伦理任务

    arXiv
    2505.23559
    Published
    Layer
    Application layer
    Setting
    science agent, Multi-agent systems
    Evaluated models
    GPT-4o, GPT-4.1, GPT-3.5-turbo and others5
    Abstract四层监控加科研安全基准。

    SafeScientist 是带四层监控的科研智能体框架，并用 SciSafetyBench 评测科学场景中的任务安全。

    Paper analysisRead analysis
You have reached the end