Skip to content

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

Oct 13 · Tuesday

All papers

Found 1 paper
Filter3

Layer

Model layer1

System component

Memory0Guardrails and judges1Monitors3Reasoning chains6Vision9Audio5Image generation3Multimodal17
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 213papers have no tags yet and are not included in filtered results.
Layer：Model layerSystem component：MemorySystem component：Guardrails and judgesClear all

Another 213papers do not have category tags yet and are not included in filtered results.

  1. InterpretabilityarXiv 2609.05074

    Influence Score 与 Transformer 可解释性：推理时注意力头有效影响的度量

    提出 influence score，量化提示注入检测器各注意力头的有效贡献

    arXiv
    2609.05074
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    DeBERTa-v3-base prompt-injection classifier
    Abstract该分数把注意力头的方向与残差贡献连到分类，并分开正确与错误预测。

    这项工作在推理时给提示注入检测分类器的注意力头一个影响分数，用来看分类决策从哪些头来。缺口是电路分析往往要重插桩，只看输出又不知道内部组件的贡献。作者希望同时看到决策在哪一层形成、哪些头在推，以及正确与错误是否对应不同机制。

    Paper analysisRead analysis
You have reached the end