Skip to content

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

Oct 13 · Tuesday

All papers

Found 1 paper
Filter5

Model risks

Misalignment and value divergence0Inappropriate refusal1

System type

Models and APIs93AI Agent4coding agent1

Layer

Model layer1Application layer4Training and data layer0
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 218papers have no tags yet and are not included in filtered results.
Model risks：Misalignment and value divergenceModel risks：Inappropriate refusalSystem type：coding agentLayer：Model layerLayer：Training and data layerClear all

Another 218papers do not have category tags yet and are not included in filtered results.

  1. InterpretabilityarXiv 2606.05396

    研究用消融移除代码 LLM 拒答方向，分离拒答与生成能力

    用 abliteration 消融代码 LLM 拒答方向，分离拒答与漏洞注入能力

    arXiv
    2606.05396
    Published
    Layer
    Model layer
    Setting
    coding agent
    Evaluated models
    Qwen2.5-Coder-Instruct 3B, Qwen2.5-Coder-Instruct 7B, Qwen2.5-Coder-Instruct 14B
    Abstract拒答可用低秩编辑去掉，CWE-89 注入是否成功仍由模型规模决定。

    这项初步可行性研究用 abliteration 把代码 LLM 的拒答和 CWE 注入能力拆开：意愿可以被低秩编辑解锁，注入是否成功仍看参数规模。

    Paper analysisRead analysis
You have reached the end