Skip to content

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

Oct 12 · Monday

All papers

Found 1 paper
Filter2

Model risks

Dangerous agent actions1

System component

Permissions and sandboxing1Multimodal3
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.
Model risks：Dangerous agent actionsSystem component：Permissions and sandboxingClear all

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

  1. Risky behaviorarXiv 2609.35799

    研究者复现 OpenAI-HuggingFace 事件中的失准行为，并提出自动化对齐测试方法

    复现共享基础设施中的级联失准行为，并提出自动化对齐测试方法

    arXiv
    2609.35799
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GLM 5.2, GLM 5.3, Kimi K3 and others9

    Abstract论文复现了 OpenAI–HF 事件四步失准行为，作者报告高阶描述结合算力可自动诱导，并展示了 RL 降本潜力。

    Paper analysisRead analysis
You have reached the end