Skip to content

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

Oct 12 · Monday

All papers

Found 5 papers
Filter5

Attack type

Detection evasion3

Attack technique

Gradient and representation optimization8Defense-targeted5Dormant triggers0

System type

Models and APIs3

Attacker access

Black-box7White-box5

Layer

Model layer3Application layer2
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 233papers have no tags yet and are not included in filtered results.
Attack technique：Defense-targetedAttack technique：Dormant triggersAttacker access：White-boxLayer：Model layerLayer：Application layerClear all

Another 233papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2610.06670

    研究者提出 Reward Stealing Attack，从对齐模型行为反推安全奖励实施攻击

    提出 ReSA，从对齐模型行为恢复安全奖励并在解码时反转

    arXiv
    2610.06670
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    White-box, Gray-box
    Evaluated models
    Llama3.1-8B-Instruct, Gemma-7B-Instruct, Qwen2.5-7B-Instruct and others9
    AbstractReSA 用最大熵 IRL 从对齐模型行为恢复代理安全奖励，并在解码时反转以诱导有害输出。

    ReSA 是一种对抗攻击：从对齐模型的可观察行为恢复代理安全奖励，再在解码时把该奖励反过来用。。

    Paper analysisRead analysis
  2. AttacksarXiv 2610.03166

    LiBRA：通过双向隐空间优化实现检测感知的图像水印去除

    提出 LiBRA，在潜空间双向优化以规避图像水印双侧检测

    arXiv
    2610.03166
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    White-box
    Evaluated models
    Stable Signature, YU1, HiDDeN and others4

    AbstractLiBRA 用双向潜空间优化规避双侧水印检测，并提高相对反转基线的保真。

    Paper analysisRead analysis
  3. 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 2609.39607

    Pretext 攻击绕过 NVIDIA SkillSpector 等 Agent 技能扫描器

    提出 PRETEXT 攻击，迭代改写技能文本绕过技能扫描器并让 Agent 执行木马

    arXiv
    2609.39607
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Attacker
    White-box
    Evaluated models
    glm (z-ai/glm-5.1), qwen3t (qwen/qwen3-235b-a22b-thinking-2507), gpt-oss (openai/gpt-oss-120b)
    Abstract论文提出了针对多阶段技能检测器的白盒攻击，揭示了单纯依赖大模型自适应检测的局限。

    PRETEXT 是一项针对基于“静态规则加语义大模型”的智能体技能安全检测框架的白盒攻击与协同进化评估研究。

    Paper analysisRead analysis
You have reached the end