Skip to content

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

Oct 13 · Tuesday

All papers

Found 6 papers
Filter6

Model risks

Dangerous agent actions3

Impact

Harmful content5Unauthorized actions3Output manipulation4Availability and reliability3Oversight failure3

System type

Models and APIs15AI Agent6coding agent0Embodied systems and robotics6

Layer

Prompt layer5Model layer6Application layer250Training and data layer11
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 215papers have no tags yet and are not included in filtered results.
Impact：Unauthorized actionsImpact：Availability and reliabilitySystem type：AI AgentSystem type：coding agentSystem type：Embodied systems and roboticsLayer：Model layerClear all

Another 215papers do not have category tags yet and are not included in filtered results.

  1. 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
  3. AttacksarXiv 2610.06083

    研究者提出面向深度强化学习的轨迹级可迁移黑盒对抗攻击

    提出轨迹级可迁移对抗攻击，降低深度强化学习策略的期望回报

    arXiv
    2610.06083
    Published
    Layer
    Model layer
    Setting
    Embodied systems and robotics
    Evaluated models
    DQN, Double DQN (DDQN)
    Abstract轨迹级后退视界攻击在 CartPole 上比逐步 FGSM 类和 Sign 噪声更压低回报。

    这项工作研究 DRL 上的迁移式黑盒观测攻击：受害参数不可访问时，在白盒代理上造扰动再迁移。

    Paper analysisRead analysis
  4. DefensesarXiv 2610.05166

    安全动作不等于可行动作：VICS-G 为 VLA 策略做可行未来解码

    提出 VICS-G 可行未来解码，重排冻结 VLA 策略的动作候选

    arXiv
    2610.05166
    Published
    Layer
    Model layer
    Setting
    Embodied systems and robotics
    Evaluated models
    SafeVLA (task-optimized / base), SafeVLA (safety-aligned / safe)
    Abstract作者用可行未来解码重排冻结 VLA 的下一步，并称 VICS-G 降低观测代价且完成率接近策略采样。

    训练无关解码器 VICS 在冻结 VLA 的推理阶段重排动作块，使当前选择带上该动作仍可能留下的安全完成质量。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2604.12447

    HazardArena：面向 VLA 模型的语义安全评测基准

    提出 HazardArena 基准，用孪生场景评测 VLA 的语义安全

    arXiv
    2604.12447
    Published
    Layer
    Model layer
    Setting
    Embodied systems and robotics
    Evaluated models
    OpenVLA-OFT, π0, NORA and others4
    Abstractsafe-only 微调提升良性执行，也提升匹配不安全孪生上的危险完成。

    HazardArena 评测 VLA 在运动要求不变时，能否因语义风险而停手。

    Paper analysisRead analysis
You have reached the end