Skip to content

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

Oct 12 · Monday

All papers

Found 35 papers
Filter4

Attack type

Jailbreaks12Prompt injection17Poisoning and backdoors8Malicious use8

Attack technique

Automated search3

Model risks

Dangerous agent actions3

Defense type

Detection and filtering4Monitoring and auditing5

Impact

Harmful content10Unauthorized actions15Data and privacy leaks4Output manipulation3

System type

Models and APIs57LLM applications5AI Agent35coding agent7web agent4Browser and computer use2Multi-agent systems5

Attacker access

Black-box3

Attack surface / entry point

User input35Web pages and interfaces15Documents and messages22Tool responses23Tool and skill descriptions13Code repositories and dependencies7Retrieval corpus4Memory writes11Inter-agent messages8Images, audio and video5Distribution channels8

Layer

Application layer33

System component

MCP and skills3Memory6Guardrails and judges7Monitors6
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 255papers have no tags yet and are not included in filtered results.
System type：AI AgentSystem type：coding agentSystem type：Browser and computer useAttack surface / entry point：User inputClear all

Another 255papers 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
  2. Evaluation and benchmarksarXiv 2610.05586

    AgentDoxx：联网搜索 Agent 对匿名访谈文本的再识别评测

    构建 AGENTDOXX 基准，评测联网搜索 Agent 对匿名访谈的再识别

    arXiv
    2610.05586
    Published
    Layer
    Application layer
    Setting
    web agent
    Evaluated models
    GPT-5.6 Terra, GPT-5 Mini, GPT-4.1 Mini and others10

    Abstract论文提出了已知身份的重识别基准 AGENTDOXX，揭示搜索与记忆双重风险及现有防御在检索轨迹中的泄露漏洞。

    Paper analysisRead analysis
  3. Evaluation and benchmarksarXiv 2610.04589

    StegoMemory red team testing: agent persistent memory can be used as a cross-session covert channel

    提出 StegoMemory，评测智能体持久记忆作为跨会话隐写信道

    arXiv
    2610.04589
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Llama 3.1 405B, Claude 3.5 Sonnet, GPT-4o and others13

    Abstract研究揭示智能体持久记忆可作为隐蔽信道跨会话传输机密，写入持久化是主要防御瓶颈，呼吁建立记忆完整性防御。

    Paper analysisRead analysis
  5. Evaluation and benchmarksarXiv 2610.03434

    Persona Guardrail：面向智能体系统的生产级防御框架

    提出 Persona Guardrail，强制客服智能体守住功能边界

    arXiv
    2610.03434
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Attacker
    Black-box
    Evaluated models
    Qwen3-30B-A3B, Qwen3-VL-30B, Qwen3-8B and others5
    AbstractPersona Guardrail 用功能边界做同步输入输出校验，并在 PAGE 与生产延迟预算下给出评测。

    Persona Guardrail 是客服向智能体的同步运行时防御：先写明功能边界，再决定放行或拦截，而不是只检测某一攻击类。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2609.39146

    MADBench：多智能体辩论安全性评测基准发布

    提出 MADBench，评测多智能体辩论是吸收还是放大对抗影响

    arXiv
    2609.39146
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    gpt-4o, GPT-5, GPT-6 Astra
    AbstractMADBench 显示辩论能吸收部分准确率攻击，同时放大未授权读写。

    MADBench 是一个评测多智能体辩论安全性的基准，用来看攻击是被辩论吸收还是被放大。

    Paper analysisRead analysis
  8. Evaluation and benchmarksarXiv 2609.33102

    ORBIT：基于 UK AISI Inspect 的多智能体安全评测框架

    提出 ORBIT 框架，评测多智能体防御对合谋等威胁的效果

    arXiv
    2609.33102
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    GPT-5.4, Claude Sonnet 4.6, Qwen3-235B and others5

    AbstractORBIT 用六层配置比较多智能体防御，逐动作监控不迁移到合谋。

    Paper analysisRead analysis
  9. Evaluation and benchmarksarXiv 2609.32616

    Cave-Bench：虚假指控下 LLM 智能体最高 60.06% 的运行会破坏正确工作

    提出 CAVE-BENCH，评测智能体在虚假指责下破坏已完成的正确工作

    arXiv
    2609.32616
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude-Sonnet-5, Claude-Opus-5, GPT-5.6-Sol and others14

    Abstract论文提出了评估智能体在虚假指责下破坏已有正确成果的 CAVE-BENCH，发现强模型易推翻已知证据，提出了门控缓解方案。

    Paper analysisRead analysis
  11. Evaluation and benchmarksarXiv 2609.27336

    CART：面向大语言模型的闭环自适应红队框架

    提出闭环自适应红队框架 CART，评测模型与 Agent 的安全失败

    arXiv
    2609.27336
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Claude Opus 4.8, GPT-5.5, Grok 4.3 and others7
    AbstractCART 把红队做成可审计的闭环搜索，在三类评测上比静态回放发现更多失败与更高平均风险。

    CART 把自动红队从一次清单式打分，改成带预算、可追溯的闭环搜索。

    Paper analysisRead analysis
  13. Evaluation and benchmarksarXiv 2609.22724

    MATE：面向移动 Agent 的策略感知安全审计方法

    提出移动 Agent 轨迹审计器 MATE，判断轨迹是否违反可编辑安全策略

    arXiv
    2609.22724
    Published
    Layer
    Application layer
    Setting
    Browser and computer use
    Attacker
    Black-box
    Evaluated models
    MATE-0.5B, MATE-1.5B, MATE-3B
    AbstractMATE 用合成轨迹微调轻量审计器，在可编辑策略下判断移动智能体轨迹是否违规并给出解释。

    MATE 是面向移动 GUI 智能体的轻量策略条件轨迹审计器，在给定自然语言安全策略时判断轨迹是否违规并解释原因。

    Paper analysisRead analysis
  16. Evaluation and benchmarksarXiv 2609.09798

    CS-Guard：首个面向代码生成安全的 LLM 护栏评测基准

    提出 CS-Guard 基准，评测代码生成护栏对虚构场景攻击的防御

    arXiv
    2609.09798
    Published
    Layer
    Prompt layer
    Setting
    coding agent
    Attacker
    Black-box
    Evaluated models
    CodeLlama-13B-Instruct, DeepSeekCoder-V2-Lite-16B, GPT-OSS-20B and others7

    Abstract论文提出了代码安全护栏基准 CS-Guard，评估发现现有护栏对虚构场景攻击和代码级恶意请求防御较弱。

    Paper analysisRead analysis
  17. Evaluation and benchmarksarXiv 2609.06966

    Mole：面向前沿实验室 AI 智能体的内部威胁检测基准

    提出 MOLE 基准，评测监控器对 AI 智能体内部威胁的检出能力

    arXiv
    2609.06966
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5.3 Chat, GPT-5.5, GPT-4.1 and others15
    AbstractMOLE 构建了智能体内部威胁检测基准，作者指出高漏检现状与推理痕迹的防御价值，并展示了监控策略优化空间。

    MOLE 是首个针对前沿 AI 实验室智能体内部威胁检测的开放基准，旨在评估在有限每日审查预算与日常研发业务背景下识别恶意账号的能力。。

    Paper analysisRead analysis
  19. Evaluation and benchmarksarXiv 2608.16246

    CompoSkill 框架利用逐个通过扫描的 Agent 技能组合发起攻击

    提出 CompoSkill，将逐个通过扫描的技能编排为攻击链

    arXiv
    2608.16246
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5.4, Gemini-3.1-flash, DeepSeek-V4 and others4

    Abstract论文分析了智能体技能组合风险，指出单包安全认证状态不具备可组合性，需转向路径级防护。

    Paper analysisRead analysis
Load more