Skip to content

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

Oct 12 · Monday

All papers

Found 2 papers
Filter3

Attack type

Jailbreaks55Prompt injection25Poisoning and backdoors15Model tampering8Extraction and theft2Detection evasion16

Attack technique

Gradient and representation optimization1Automated search6Encoding and obfuscation1Role-play and persuasion3Side channels4
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：Encoding and obfuscationClear all

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

  1. AttacksarXiv 2604.01039

    评估与加固 LLM 系统指令以抵御编码攻击提取

    用编码与结构化请求抽取系统指令，并用 CoT 改写加固

    arXiv
    2604.01039
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    GPT-4.1-mini, GPT-3.5-turbo, Gemini-2.5-flash and others7
    Abstract编码式请求可在直接拒答后引出系统指令。

    这项工作评测并加固 LLM 系统指令在编码式抽取下的保密性。作者关心的是：直接抽取已被拒绝时，把同一请求改写成格式化或编码任务，模型是否仍交出受保护内容。

    Paper analysisRead analysis
  2. 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
You have reached the end