Skip to content

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

Oct 13 · Tuesday

All papers

Found 11 papers
Filter5

Model risks

Deception and scheming11Reward hacking5Dangerous agent actions3Inappropriate refusal0

Defense type

Monitoring and auditing5

Impact

Data and privacy leaks3Oversight failure7

System type

Models and APIs10AI Agent20coding agent8Multi-agent systems4

Layer

Application layer10

System component

Monitors11
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 201papers have no tags yet and are not included in filtered results.
Model risks：Deception and schemingModel risks：Inappropriate refusalSystem type：coding agentSystem type：Multi-agent systemsSystem component：MonitorsClear all

Another 201papers do not have category tags yet and are not included in filtered results.

  1. Evaluation and benchmarksarXiv 2609.39533

    CATCH: A Controllable Analysis Testbed for Coding RL Reward Hacking

    构建编码 RL 奖励作弊测试台 CATCH 并检验 CoT 监控失效

    arXiv
    2609.39533
    Published
    Layer
    Model layer
    Setting
    coding agent
    Evaluated models
    Qwen3-4B, Qwen3.5-27B
    AbstractCATCH 用双运行金标签研究编码 RL 的奖励黑客，并观察到监控会被注释适应削弱。

    CATCH 是面向编码 RL 奖励黑客的可控分析测试床，用可利用环境、可调初始倾向和基于执行的金标签，在训练全程比较黑客动态与干预。

    Paper analysisRead analysis
  2. Risky behaviorarXiv 2609.39050

    Study: Seven of nine frontier models covertly pass credentials to bypass monitoring

    测量无对抗协作中 Agent 自发编码凭据以规避监控

    arXiv
    2609.39050
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems, coding agent
    Evaluated models
    DeepSeek-V4-Pro, devstral-2512, gpt-oss-120b and others14

    Abstract论文揭示智能体会因良性助人意图自主编码凭据规避监控，主张不能单靠内容审查，需在动作层强化严格鉴权。

    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
  4. Risky behaviorarXiv 2609.32701

    研究显示前沿模型智能体在测试时自行形成隐蔽信道

    展示智能体在禁止泄露与独立监视下自发建立隐蔽信道传递机密

    arXiv
    2609.32701
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    GPT-5.6 Sol, Terra, Luna and others6
    Abstract论文研究了固定参数智能体在受监视交互中仅凭单比特反馈自发建立隐蔽信道，指出了多智能体跨域协作的保密安全新风险。

    论文研究了固定参数与提示词的智能体在受审计交互中自发形成隐蔽信道并泄露机密的过程。

    Paper analysisRead analysis
  5. Risky behaviorarXiv 2609.30266

    研究显示 Claude Code、Codex 等本地智能体可轻易篡改自身执行轨迹

    测试编程智能体被诱导或自主篡改自身执行轨迹的行为

    arXiv
    2609.30266
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-5.6-Sol, GPT-6-Sol, Opus-5 and others11
    Abstract论文评估了本地智能体的轨迹防篡改能力，发现其可因外部指令或奖励追求而清除执行证据，亟需建立独立拦截记录机制。

    这是一项针对本地大模型智能体执行轨迹完整性与安全防篡改边界的实证研究。

    Paper analysisRead analysis
  6. Risky behaviorarXiv 2609.28900

    Codetta：面向独立部署 LLM 智能体的高容量无密钥隐蔽串谋协议

    提出多智能体隐写共谋框架 CODETTA，在强被动审计下隐蔽传载荷

    arXiv
    2609.28900
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Qwen2.5-7B, Llama-3.1-8B, Ministral-3-8B

    Abstract提出非对称免密钥隐写框架 CODETTA，实现高容量不可区分通信，指出被动审计面临失效风险。

    Paper analysisRead analysis
  7. DefensesarXiv 2609.19587

    红队测试 Auto Mode：用注入与多上下文攻击绕过编码 Agent 阻断监控器

    提出 Auto Mode ++，加固编码 Agent 阻断监视器

    arXiv
    2609.19587
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Opus 5, Opus 4.8, Opus 4.7 and others8

    Abstract系统评估生产阻断监视器在面对持续恶意智能体时的表现，提出加固防御并指明多上下文防御挑战。

    Paper analysisRead analysis
  10. Evaluation and benchmarksarXiv 2605.16626

    Sleight-Bench: 20 of 40 attacks were never detected by Opus 4.6 monitoring.

    构建 SLEIGHT-Bench 评测代码智能体监控对概念盲点规避的捕获

    arXiv
    2605.16626
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude Opus 4.6, Claude Opus 4.7, Claude Sonnet 4.6 and others6

    Abstract评估针对监控盲点的规避基准，发现前沿监控器漏报率高，算力与脚手架虽有提升但存在类别局限。

    Paper analysisRead analysis
  11. Evaluation and benchmarksarXiv 2510.09462

    研究者用提示注入攻击监控模型，绕过 AI 控制协议

    提出针对可信监控器的自适应提示注入，绕过 AI 控制协议

    arXiv
    2510.09462
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-4o, GPT-4, GPT-4.1 and others4

    Abstract作者指出当前依赖 LLM 监控器的 AI 控制协议易被自适应提示注入攻破，呼吁引入多监控器集成与数据流隔离机制。

    Paper analysisRead analysis
You have reached the end