Skip to content

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

Oct 12 · Monday

All papers

Found 46 papers
Filter2

Attack type

Jailbreaks14Prompt injection5Poisoning and backdoors19Extraction and theft7Detection evasion4

Attack technique

Automated search14Encoding and obfuscation9Cross-modal payloads5Defense-targeted4Dormant triggers7

Impact

Harmful content14Data and privacy leaks9Output manipulation22

System type

Models and APIs22LLM applications19AI Agent5

Attacker access

Black-box17

Attack surface / entry point

User input23Retrieval corpus17Images, audio and video8Weights and gradients5Distribution channels3

Layer

Prompt layer15Model layer4Application layer24Training and data layer3

System component

MCP and skills38Memory21RAG24Guardrails and judges55Monitors6Reasoning chains21Permissions and sandboxing12Fine-tuning and open weights27Vision55Audio13Video3Image generation22Embeddings8Multimodal91
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.
System component：RAGSystem component：Image generationClear all

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

  1. AttacksarXiv 2610.11932

    A Fudan team measures the citation fragility of AI search: ordinary posts can enter citations and answer text

    Measures citation fragility in AI search, demonstrating that ordinary public posts can enter citations and answers.

    arXiv
    2610.11932
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    Grok, Doubao, ChatGPT and others10

    AbstractThe authors claim the paper quantifies vulnerabilities in AI search caused by citation preferences and low posting thresholds; tests show public web posts can penetrate AI citations, and they call for auditing the citation supply chain.

    Paper analysisRead analysis
  2. AttacksarXiv 2610.11112

    EpiReal-Bench：商用图像生成模型虚假声明红队基准

    构建 EpiReal-Bench，评测图像生成器将虚假主张渲染为可信证据

    arXiv
    2610.11112
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    GPT-Image-2, Nano Banana 2, Grok Imagen 2.0 and others4
    Abstract四款商用图像模型能把假主张画成可信证据。

    EpiReal-Bench 用来量商用图像生成器会不会把虚假现实主张画成可信视觉证据。

    Paper analysisRead analysis
  4. AttacksarXiv 2610.05601

    Tracer 框架可从遗忘后的扩散模型中识别被擦除概念

    提出 TRACER，从权重差识别扩散模型被擦除概念

    arXiv
    2610.05601
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    White-box
    Evaluated models
    Stable Diffusion v1.5, Stable Diffusion 3, FLUX.1 and others5
    AbstractTRACER 用权重足迹识别未披露擦除概念并估计个数，作者称数秒内完成且准确率高于对照。

    TRACER 针对的是遗忘扩散模型上的目标识别：提供者从公开基座 W 用已发表方法擦掉未知集合 S⋆，只发布 W′。攻击者还有假定包含这些概念的词表，但不知擦了谁、擦了几个、用了什么算法。

    Paper analysisRead analysis
  5. 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
  9. AttacksarXiv 2609.38253

    CollageAttack: Exploiting cross-modal alignment defects for spatial text composition jailbreaks against T2I models.

    提出 CollageAttack，用空间文本碎片重组越狱文生图模型

    arXiv
    2609.38253
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    GPT-Image-2, Doubao-Seedream-5.0-Lite, Wan2.7-Image-Pro and others5

    Abstract系统揭示 T2I 跨模态空间合成安全漏洞，证实离散无害元素可在图像层面重构有害意图并放大传播强度。

    Paper analysisRead analysis
  13. AttacksarXiv 2609.37468

    论文提出针对 Agentic RAG 的检索器后门攻击与 DOU 隐藏方法

    提出检索器后门攻击操纵智能体搜索，并用 DOU 逃避检测

    arXiv
    2609.37468
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    E5, Contriever-MSMARCO

    Abstract论文展示了通过恶意检索器操纵智能体多轮搜索并利用诱饵重塑表征隐蔽后门的可行性与系统风险。

    Paper analysisRead analysis
  15. AttacksarXiv 2609.14649

    CIG-MIA：利用上下文信息增益对 RAG 系统发起成员推断攻击

    提出 CIG-MIA，用上下文信息增益推断 RAG 知识库成员身份

    arXiv
    2609.14649
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Attacker
    Gray-box, Black-box
    Evaluated models
    Llama-3.1-8B-Instruct, Mistral-7B-Instruct-v0.3, Llama-2-13b-chat-hf and others6
    AbstractCIG-MIA 用注入前后的答案似然差推断 RAG 库成员，灰盒可分性高于所列基线。

    CIG-MIA 是针对 RAG 外部知识库的成员推断攻击，同时覆盖灰盒和纯文本黑盒。

    Paper analysisRead analysis
  17. AttacksarXiv 2609.12448

    GraphProfiler 用个人知识图谱推断敏感属性并溯源到具体帖子

    提出 GraphProfiler，用个人知识图谱从帖子历史推断敏感属性并溯源

    arXiv
    2609.12448
    Published
    Layer
    Application layer
    Setting
    LLM applications
    AbstractGraphProfiler 用源链接个人知识图谱做属性推断，准确率接近文本基线，并把证据定位到少量源帖。

    GraphProfiler 是一个可审计的 LLM 敏感属性画像器：用源链接个人知识图谱，把推断落到具体帖子。

    Paper analysisRead analysis
  19. AttacksarXiv 2609.07216

    DIVA：利用视觉锚点对视频生成模型实施多模态越狱

    提出 DIVA，将有害意图拆成视觉锚点与运动文本以越狱视频生成模型

    arXiv
    2609.07216
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Wan, CogVideoX, LTX-Video and others9
    AbstractDIVA 用视觉锚点抑制语义漂移做视频越狱，并给出多条件安全基准 TI2VSafetyBench。

    DIVA 是针对带参考图的视频生成模型的免训练 black-box 越狱：用一致性机制本身堵住语义漂移这条安全逃逸路径。。

    Paper analysisRead analysis
Load more