Skip to content

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

Oct 13 · Tuesday

All papers

Found 5 papers
Filter4

Attack type

Jailbreaks5Prompt injection48Poisoning and backdoors11Model tampering2Extraction and theft3Malicious use3

Impact

Harmful content3

System type

Models and APIs48AI Agent14coding agent5web agent2Multi-agent systems3

Attack surface / entry point

User input4

Layer

Application layer5
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 227papers have no tags yet and are not included in filtered results.
Attack type：Model tamperingAttack type：Malicious useSystem type：web agentSystem type：Multi-agent systemsClear all

Another 227papers do not have category tags yet and are not included in filtered results.

  1. 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
  2. AttacksarXiv 2609.39788

    研究：多智能体系统潜在通信链路可被训练用于提升有害顺从

    提出只训练或篡改多智能体潜在通信链路来提高有害遵从

    arXiv
    2609.39788
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Llama-3.2-3B-Instruct, Qwen2.5-3B-Instruct, Gemma-3-1B-IT and others7
    Abstract潜在链路本身即可改变冻结智能体系统的安全，奖励既能攻击也能修复。

    学习得到的潜在通信链路会改变由冻结、对齐智能体组成的系统是否遵从有害请求。。作者研究三种拓扑：双智能体、三级顺序协作，以及双专家加汇总。

    Paper analysisRead analysis
  3. 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
  5. Evaluation and benchmarksarXiv 2503.04957

    SafeArena：首个评估自主网页 Agent 滥用风险的基准

    提出 SafeArena 基准，评测网页 Agent 是否执行故意有害任务

    arXiv
    2503.04957
    Published
    Layer
    Application layer
    Setting
    web agent
    Evaluated models
    GPT-4o, Claude-3.5-Sonnet, Qwen-2-VL-72B and others5
    AbstractSafeArena 用配对任务和 ARIA 评测网页智能体的故意滥用。

    SafeArena 评测自主网页智能体是否会执行用户故意给出的恶意网页任务。。

    Paper analysisRead analysis
You have reached the end