Skip to content

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

Oct 12 · Monday

All papers

Found 14 papers
Filter3

Attack type

Prompt injection13Poisoning and backdoors6

Defense type

Model hardening6Detection and filtering14Instruction–data separation23Permissions and approval21Monitoring and auditing14

Impact

Unauthorized actions9Output manipulation4

System type

AI Agent13

Attack surface / entry point

User input16Web pages and interfaces8Documents and messages10Tool responses14Tool and skill descriptions10Retrieval corpus5Inter-agent messages3Images, audio and video5Distribution channels7

Layer

Application layer14

System component

Memory3Guardrails and judges9
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 265papers have no tags yet and are not included in filtered results.
Defense type：Detection and filteringAttack surface / entry point：Tool responsesLayer：Application layerClear all

Another 265papers do not have category tags yet and are not included in filtered results.

  1. DefensesarXiv 2610.07657

    DEFER1 多智能体防御研究报告受控测试中的攻击成功率下降

    实现 DEFER，以确定性检查优先拦截多智能体流水线的内容攻击

    arXiv
    2610.07657
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Qwen3-235B-A22B-Instruct-2507, gpt-oss-120b, Llama-4-Scout-17B-16E-Instruct and others5
    AbstractDEFER 把确定性检查放在 judge 之前。

    DEFER 用来测量多智能体授权流水线里，确定性规则在何处结束、LLM judge 从何处开始。

    Paper analysisRead analysis
  2. Evaluation and benchmarksarXiv 2610.03448

    研究重测 15 个提示注入检测器：基准分数难以预测 Agent 部署表现

    重测提示注入检测器在 Agent 工具输出上的检出与任务阻断

    arXiv
    2610.03448
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Horizon-Labs, Wolf Defender, Prismor-1.5B and others15

    Abstract论文揭示公开基准无法反映智能体检测器表现，检测能力源于输入格式相似而非通用防御，建议基于工具输出与低误报评测。

    Paper analysisRead analysis
  3. Evaluation and benchmarksarXiv 2609.35886

    Agentic Commerce Bench：衡量会花钱的智能体的欺诈检测能力

    构建支付欺诈基准 ACB，评测花钱智能体的欺诈检测

    arXiv
    2609.35886
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    gordonguard, pattern judge
    AbstractACB 把支付欺诈与智能体操纵分开测量，作者强调观察权限、误报预算和微支付复核成本。

    ACB 是一套智能体支付欺诈基准与检测参考栈，目标是衡量钱是否被不当提取，而不仅是智能体是否接受了恶意指令。

    Paper analysisRead analysis
  4. DefensesarXiv 2609.33039

    TACIT：从 LLM 内部状态检测 Agent 轨迹危害

    提出 TACIT，从冻结内部状态检测智能体有害轨迹

    arXiv
    2609.33039
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Qwen3-4B, Llama-3.1-8B, Qwen3-4B-Instruct and others4

    Abstract提出基于内部表征的智能体轨迹安全检测方法 TACIT，揭示工具风险表征特性并在多项基准上提升检测表现与效率。

    Paper analysisRead analysis
  12. DefensesarXiv 2603.14707

    视觉混淆代理：利用与防御计算机使用智能体的感知失效

    提出进程外双通道对比分类，拦截计算机使用智能体的错误点击

    arXiv
    2603.14707
    Published
    Layer
    Application layer
    Setting
    Browser and computer use
    Evaluated models
    multi-modal-embed-small
    Abstract感知错误可被屏幕像素对调利用。

    作者把 CUA 的感知失败写成安全漏洞visual confused deputy：智能体按看错的屏幕授权 click(x,y)，真实落点与它以为的元素不一致。

    Paper analysisRead analysis
  13. DefensesarXiv 2601.10156

    ToolSafe 提出步骤级护栏 TS-Guard 与 TS-Flow，降低 Agent 有害工具调用

    提出 TS-Guard 与 TS-Flow，在执行前检测并引导 Agent 修正有害工具调用

    arXiv
    2601.10156
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-4o, TS-Guard, Qwen2.5-14B-Instruct
    AbstractTS-Bench 标步骤级风险，TS-Guard 用多任务 RL 做执行前判定，TS-Flow 用反馈代替直接中止。

    ToolSafe 研究 LLM 智能体在工具执行前的步骤级安全：基准 TS-Bench、护栏 TS-Guard，以及反馈驱动框架 TS-Flow。

    Paper analysisRead analysis
You have reached the end