Skip to content

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

Oct 12 · Monday

All papers

Found 4 papers
Filter3

Attack type

Jailbreaks4Extraction and theft8Detection evasion4

Attack technique

Defense-targeted3

Impact

Output manipulation3

System type

Models and APIs4

Attacker access

Black-box4White-box3

Layer

Prompt layer8Model layer4Application layer9
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.
Attack type：Detection evasionAttacker access：Black-boxLayer：Model layerClear all

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

  1. AttacksarXiv 2610.03124

    研究提出 Untag 攻击框架，使开源大模型触发标签滥用检测机制失效

    提出 UNTAG，绕过开源权重模型的 trigger-tag 滥用检测

    arXiv
    2610.03124
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    Black-box, White-box
    Evaluated models
    Llama2 (7B), Llama3 (8B), Qwen2.5 (7B)
    Abstract作者用 UNTAG 评估 trigger-tag，认为其不能充当开放权重滥用检测的主要边界。

    UNTAG 是针对开放权重 LLM 上条件滥用检测 trigger-tag 的统一攻击评估：信号可以在解码期注入，也可以写进权重，但攻击者若能改输出或改已发布权重，检测可以失效。

    Paper analysisRead analysis
  4. AttacksarXiv 2510.10987

    DITTO：通过知识蒸馏伪造水印 LLM 的作者归属

    提出 DITTO 框架，蒸馏提取水印信号并注入外部模型以伪造归属

    arXiv
    2510.10987
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Llama3.1-8B, Llama3.2-3B, GPT-OSS 20B and others4

    Abstract论文针对大模型水印归属假设的脆弱性，提出黑盒仿冒攻击 DITTO，实现跨架构和采样方案的高检测率与低困惑度仿冒。

    Paper analysisRead analysis
You have reached the end