Skip to content

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

Oct 12 · Monday

All papers

Found 6 papers
Filter4

Attack type

Poisoning and backdoors3

Attack technique

Gradient and representation optimization3Automated search8Encoding and obfuscation7Cross-modal payloads5Defense-targeted3Dormant triggers3

Impact

Output manipulation4

System type

Models and APIs5

Attacker access

Black-box3

Attack surface / entry point

Images, audio and video3

System component

MCP and skills16Memory25RAG5Guardrails and judges19Monitors5Reasoning chains5Permissions and sandboxing4Fine-tuning and open weights24Vision3Image generation5Embeddings1Multimodal11
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 triggersSystem component：Image generationSystem component：EmbeddingsClear all

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

  1. AttacksarXiv 2610.05453

    研究者提出 Privacy-Leaking Watermarks，可在本地多模态模型中泄露对话隐私

    提出 PLW，用微调把先前对话敏感触发编码进后续生成图像

    arXiv
    2610.05453
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    OmniGen2, BAGEL-7B

    AbstractPLW 把先前对话线索写成后续图像中的隐水印，检出随触发、模型和间隔变化。

    Paper analysisRead analysis
  6. AttacksarXiv 2512.01335

    EmoRAG 研究揭示 RAG 检索对表情符号扰动的脆弱性

    提出 EmoRAG，向查询与知识库注入同一表情以劫持 RAG 检索

    arXiv
    2512.01335
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    SPECTER, Contriever, Qwen2-7B and others11
    AbstractEmoRAG 用表情联动查询与库文本。

    EmoRAG 测量 RAG 对符号扰动的反应：同一表情同时出现在查询和库文本里时，检索会偏离语义匹配。作者要检查的是，检索是否真由语义相关性决定。

    Paper analysisRead analysis
You have reached the end