Skip to content

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

Oct 13 · Tuesday

All papers

Found 2 papers
Filter3

Attack type

Model tampering2Extraction and theft0

Attack surface / entry point

Training data2
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.
Attack type：Model tamperingAttack type：Extraction and theftAttack surface / entry point：Training dataClear all

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

  1. InterpretabilityarXiv 2610.07518

    TRACE：仅凭 checkpoint 更新审计大语言模型有害目标

    提出 TRACE，仅凭检查点更新审计有害合规微调目标

    arXiv
    2610.07518
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Llama-3.1-8B-Instruct, Mistral-7B-Instruct-v0.3, Qwen3-8B and others6
    Abstract作者称有害合规 SFT 在检查点更新中留下可排序坐标，并据此做只读权重审计。

    作者在检查点更新里寻找有害合规 SFT 的目标组成，并把读数做成不查询模型、不看未知训练数据的审计分 TRACE。

    Paper analysisRead analysis
  2. InterpretabilityarXiv 2609.06934

    论文提出拒答的几何解释：事后安全训练脆弱，预训练期持续安全共训更抗攻击

    解释后验安全的权重几何脆弱性，并提出预训练持续安全共训练

    arXiv
    2609.06934
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen-2.5-7B, Qwen-2.5-7B-Instruct, Llama-3-8B and others15

    Abstract论文将后验安全脆弱性归因于正交抑制，并提出通过预训练持续共训练实现具备微调抗性的安全防护。

    Paper analysisRead analysis
You have reached the end