Skip to content

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

Oct 12 · Monday

All papers

Found 5 papers
Filter1

Impact

Harmful content17Unauthorized actions3Output manipulation6Oversight failure5

System type

Models and APIs5

Layer

Model layer4
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 305papers have no tags yet and are not included in filtered results.
Impact：Oversight failureClear all

Another 305papers do not have category tags yet and are not included in filtered results.

  1. InterpretabilityarXiv 2610.12361

    Study: legal reasoning models citing legal provisions does not equal relying on legal provisions.

    Uses premise permutation and hidden-layer tracing to audit whether legal chain-of-thought truly relies on the cited legal provisions.

    arXiv
    2610.12361
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-8B, Qwen3-14B, Llama-3.1-8B and others7
    AbstractUsing counterfactual substitution of authorities and internal commitment tracing, the authors find that legal CoT rulings depend on authorities only to a limited extent and are vulnerable to prompt injection manipulation.

    This study conducts an empirical audit of the causal truthfulness of legal citations in chain-of-thought reasoning of large language models in the legal domain.

    Paper analysisRead analysis
  3. InterpretabilityarXiv 2609.38081

    Hessian Null Space Continuation：在单个神经网络解附近发现大量表征多样性

    提出 HNC，在单个解附近遍历保函数权重并比较内部表示

    arXiv
    2609.38081
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    64-unit tanh RNN, ViT-S/16, RNN policy (Plume Tracking) and others5
    AbstractHNC 在保持函数的局部权重区域里找到表示、动力学或行为不同的解。

    Hessian Null Space Continuation（HNC）用局部曲率，在单个已训练网络周围遍历近似保持输入输出的连通低损失区域，并可用可微目标做引导。作者要回答的是：这个区域里是否真有不同的内部机制，而不只是同一表示的小改动。做法是交替执行零空间内的平坦步与函数匹配损失上的梯度恢复。

    Paper analysisRead analysis
  5. InterpretabilityarXiv 2608.29461

    因果模型定位并解锁模型中的故意藏拙行为

    提出因果模型定位残差流中的藏拙轴，并用单层 graft 在推理时解锁

    arXiv
    2608.29461
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen2.5-7B, Llama-3-8B, Mistral-7B
    Abstract因果模型规定单层 graft 的窗口。

    作者为故意造出的 sandbagging 建立 residual stream 上的因果模型，并用两种推理时、不改权重的 graft 检验其预测。

    Paper analysisRead analysis
You have reached the end