Skip to content

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

Oct 12 · Monday

All papers

Found 149 papers
Filter3

Attack type

Jailbreaks4Prompt injection27Poisoning and backdoors16Extraction and theft3

Attack technique

Gradient and representation optimization8Automated search4Cross-modal payloads3Dormant triggers7

Model risks

Deception and scheming15Reward hacking10Misalignment and value divergence9Dangerous agent actions15Dangerous capabilities9Inappropriate refusal5Hallucination and fabrication5

Defense type

Model hardening8Inference-time intervention4Detection and filtering5Instruction–data separation6Permissions and approval11Monitoring and auditing5

Impact

Harmful content11Unauthorized actions27Data and privacy leaks7Output manipulation37Availability and reliability14Oversight failure13

System type

Models and APIs864LLM applications136AI Agent649coding agent136science agent32web agent34Browser and computer use19Multi-agent systems78Embodied systems and robotics47

Attack surface / entry point

User input18Web pages and interfaces3Documents and messages6Tool responses6Tool and skill descriptions3Retrieval corpus3Memory writes7Inter-agent messages24Images, audio and video8Training data5Weights and gradients3Distribution channels4

Layer

Prompt layer6Model layer11Application layer116Training and data layer16

System component

MCP and skills4Memory12Guardrails and judges10Monitors12Reasoning chains3Permissions and sandboxing14Fine-tuning and open weights3Vision11Multimodal13
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.
System type：science agentSystem type：Multi-agent systemsSystem type：Embodied systems and roboticsClear all

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

  1. OtherarXiv 2610.13479

    A local agentic AI framework generates edge-case scenarios for safety assurance

    Proposes a local dual-model closed loop to generate edge scenarios for autonomous system safety assurance

    arXiv
    2610.13479
    Published
    Layer
    Application layer
    Setting
    LLM applications, Embodied systems and robotics
    Evaluated models
    PPO
    AbstractLocal closed-loop generation of JSON edge cases.

    The authors built a local agentic loop that runs on a laptop to generate edge cases for safety arguments of autonomous systems, rather than releasing a new public test set.

    Paper analysisRead analysis
  3. OtherarXiv 2610.13372

    Toward Safe World Models: Risks, Defenses, and Evaluation

    A survey of risks, defenses, and evaluation in the six-stage lifecycle of world models

    arXiv
    2610.13372
    Published
    Layer
    Application layer
    Setting
    Embodied systems and robotics
    AbstractSafety is written as a property of the prediction-decision closed loop; claims must be linked to observed consequences and versions.

    World model safety is organized by the nature of the prediction-decision loop, not by isolated prediction scores.

    Paper analysisRead analysis
  5. Risky behaviorarXiv 2610.12822

    Reward hacking as a transmissible strategy in multi-agent coding systems

    Show that reward hacking can propagate through code, memory, and plans in multi-agent coding systems

    arXiv
    2610.12822
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems, coding agent
    Evaluated models
    Qwen2.5-Coder-14B, Gemma-3-12B, Qwen2.5-Coder-7B and others16
    AbstractThe authors state that reward hacking can spread across models and affect routing and aggregation, and that review can reduce delivery.

    Reward hacking induced by coding RL can leave the source agent and enter subsequent collaboration and system coordination.

    Paper analysisRead analysis
  6. Analysis and theoryarXiv 2610.12436

    The study proposes an ecological theory of AI agent populations: collaboration brings a critical scale for takeoff.

    An ecological model is established for a population of mismatched agents, showing that collaboration produces a critical takeoff scale.

    arXiv
    2610.12436
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    AbstractThe authors say that collaboration changes takeoff from an individual capability threshold to a population threshold, so small-population evaluations are insufficient.

    The authors use the ecological strong Allee effect to discuss whether a mismatched agent population can take off when individual capabilities remain unchanged..

    Paper analysisRead analysis
  7. DefensesarXiv 2610.12432

    FAITH: A feasibility-aware safety filtering reinforcement learning framework for high-dimensional systems

    Proposes the FAITH safe filtering reinforcement learning framework, training high-dimensional systems to optimize task reward under constraints

    arXiv
    2610.12432
    Published
    Layer
    Training and data layer
    Setting
    Embodied systems and robotics
    AbstractFAITH separates safety filtering from reward optimization.

    FAITH is a model-free safe filtering reinforcement learning framework for high-dimensional systems: safety correction and task reward reside in separate modules.

    Paper analysisRead analysis
  9. DefensesarXiv 2610.11754

    仅凭输出审查无法验证什么：面向研究智能体的研究契约

    提出研究合同，把已批准选择绑定到执行证据并由检查器核验

    arXiv
    2610.11754
    Published
    Layer
    Application layer
    Setting
    science agent
    Evaluated models
    OpenAI gpt-5.6-sol, OpenAI gpt-5.6-terra, OpenAI gpt-5.6-luna and others18
    Abstract研究合同暴露四类输出单独建不立的关系。

    研究合同用来核验计算科学研究智能体是否仍在已批准研究之内，不判断科学是否为真。。

    Paper analysisRead analysis
  12. OtherarXiv 2610.10064

    研究者提出理解审计机制，以缓解自动化 AI 研究带来的风险

    提出 comprehension audits，以人类理解证据作为继续研发门槛

    arXiv
    2610.10064
    Published
    Layer
    Training and data layer
    Setting
    science agent
    Abstract理解审计把出示人类理解设为继续研发的门禁，并报告开源评审评论率下降。

    本文是一份研发过程保障提案：用 comprehension audits 检验责任人是否理解 AI 参与产生的研发贡献，未达标就暂停后续开发与依赖该贡献的发布。。

    Paper analysisRead analysis
  14. Analysis and theoryarXiv 2610.09517

    审计发现 CLIP 安全分数主要反映画面与提示词场景的相似度

    开环审计冻结视觉语言安全分，发现其反映场景模板而非危险感知

    arXiv
    2610.09517
    Published
    Layer
    Model layer
    Setting
    Embodied systems and robotics
    Evaluated models
    CLIP ViT-B/32, CLIP ViT-L/14, Qwen2-VL-7B
    Abstract论文审计了冻结 VLM 安全评分机制，指出其反映场景模板而非危险感知，为安全强化学习评估提供了严谨基准。

    本文是一项针对安全强化学习中冻结视觉语言模型（VLM）安全评分机制的实证审计研究。

    Paper analysisRead analysis
  15. DefensesarXiv 2610.09496

    SOUL：用稀疏特征策略遗忘缓解 VLA 模型的状态幻觉

    提出 SOUL 稀疏特征策略遗忘，缓解 VLA 状态幻觉

    arXiv
    2610.09496
    Published
    Layer
    Training and data layer
    Setting
    Embodied systems and robotics
    Evaluated models
    OpenVLA, π0.5
    AbstractSOUL 以稀疏特征遗忘 VLA 状态幻觉。

    SOUL用稀疏特征引导的策略 unlearning 处理 VLA 的 state hallucination。

    Paper analysisRead analysis
  18. OtherarXiv 2610.08917

    CARE：为视觉-语言-动作模型推理加速提供认证式选择

    提出 CARE，离线认证 VLA 加速配置的回合级失败风险并选出最快合格候选

    arXiv
    2610.08917
    Published
    Layer
    Model layer
    Setting
    Embodied systems and robotics
    Evaluated models
    OpenVLA-OFT, π0.5, π0 and others5
    AbstractCARE 以配对回合认证加速诱发失败，在预算内选最快候选，不足则保留参考。

    CARE 在部署前从加速配置中选出最快、且加速诱发失败风险能被有限样本认证为低于用户预算的候选。证据不足就保留全计算参考策略。

    Paper analysisRead analysis
  19. DefensesarXiv 2610.08761

    VeriFine：通过扩展验证实现具身推理的自我改进

    提出 VeriFine，用双环协同扩展验证以实现具身推理自我改进

    arXiv
    2610.08761
    Published
    Layer
    Training and data layer
    Setting
    Embodied systems and robotics
    Evaluated models
    Alpamayo 1.5 8B, Qwen3-VL 2B, Claude Opus 5
    AbstractVeriFine 用双环协同演化具身推理的策略、课程和评审，驾驶与导航的策略分和评审对齐都上升。

    VeriFine 把驾驶和导航推理的自改进写成策略、课程和无参考评审的协同演化，用来处理固定评审跟不上新失败模式的问题。

    Paper analysisRead analysis
  20. 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
Load more