Skip to content

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

Oct 12 · Monday

All papers

Found 7 papers
Filter3

Attack type

Jailbreaks65Prompt injection42Poisoning and backdoors28Model tampering8Extraction and theft7Detection evasion21Denial of service3

Attack technique

Gradient and representation optimization1Automated search6Role-play and persuasion3Side channels4

Impact

Data and privacy leaks7

System type

Models and APIs4AI Agent3

Attacker access

Black-box4

Attack surface / entry point

User input4

Layer

Application layer3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 261papers have no tags yet and are not included in filtered results.
Attack type：Extraction and theftAttack technique：Gradient and representation optimizationAttack technique：Automated searchClear all

Another 261papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2608.04192

    研究者提出 SkillClone，仅凭正常提问即可克隆 LLM Agent 隐藏技能功能

    提出 SkillClone，仅凭良性任务交互重构闭源 Agent 技能

    arXiv
    2608.04192
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Attacker
    Black-box
    Evaluated models
    DeepSeek-Flash, DeepSeek-Pro, GLM-5.1 and others5
    Abstract论文揭示了仅靠良性任务交互即可重构闭源智能体技能隐藏功能的风险，表明功能保密需要限制累积信息泄露。

    论文从功能保密视角系统研究了大语言模型智能体技能的行为重构威胁。

    Paper analysisRead analysis
  2. AttacksarXiv 2607.25560

    SigLeak 可从执行轨迹推断专有 Agent 技能

    提出 SigLeak，从黑盒 Agent 执行轨迹推断专有技能

    arXiv
    2607.25560
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Attacker
    Black-box
    Evaluated models
    GPT-5.4-mini, GPT-5.5, GLM-5.2 and others4

    Abstract提出了通过良性查询和对照轨迹推断黑盒专有技能的 SigLeak 框架，揭示了智能体执行可见性带来的行为侧信道泄露风险。

    Paper analysisRead analysis
  3. AttacksarXiv 2603.11132

    WebWeaver：以隐蔽的上下文推断攻破 LLM 多智能体系统拓扑机密性

    提出 WebWeaver ，只攻陷单个智能体并从对话上下文恢复多智能体通信拓扑

    arXiv
    2603.11132
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Llama 3.1 8B, Qwen 2.5 7B, Mistral NeMo 12B and others6
    AbstractWebWeaver 用上下文推断拓扑。

    WebWeaver 是针对 LLM-MAS 通信拓扑机密性的推断攻击：只控制一个任意智能体，从对话上下文恢复全图，而不是查询智能体 ID。

    Paper analysisRead analysis
  4. AttacksarXiv 2602.01757

    Zero2Text: Training-Free Cross-Domain Text Embedding Inversion Attack

    提出 Zero2Text，无需训练地跨域反演文本嵌入

    arXiv
    2602.01757
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    GTR-Base, Qwen3-Embedding-0.6B, Text-Embedding-3-small and others4
    Abstract零训练在线对齐可在未知域、无泄漏对时从黑盒 embedding 还原句子。

    Zero2Text 是一种不训练 decoder、也不使用泄漏文本–向量对的文本 embedding 反演，针对 RAG 向量库在严格黑盒、目标域未知时的还原。

    Paper analysisRead analysis
  7. AttacksarXiv 2508.06837

    Prometheus：针对文生图扩散模型的免训练提示词窃取攻击

    提出 Prometheus，从展示图反推文生图模型的在售提示词

    arXiv
    2508.06837
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    No-box (offline transfer)
    Evaluated models
    Midjourney, Leonardo.ai, DALL·E and others4
    Abstract训练无关的 proxy 搜索攻击。

    Prometheus 是针对文生图市场 showcase 的训练无关提示词窃取：攻击者看不到在售文本，只用本地开源 proxy 恢复 subject 和 modifier，并希望换 subject 后风格仍在。

    Paper analysisRead analysis
You have reached the end