Skip to content

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

Oct 12 · Monday

All papers

Found 204 papers
Filter4

Attack type

Jailbreaks38Prompt injection32Poisoning and backdoors31Extraction and theft12Detection evasion16Malicious use6

Attack technique

Gradient and representation optimization32Automated search18Multi-turn escalation5Encoding and obfuscation17Cross-modal payloads52Dormant triggers9

Model risks

Reward hacking3Dangerous agent actions3Inappropriate refusal8Hallucination and fabrication13

Defense type

Model hardening14Inference-time intervention10Detection and filtering20Instruction–data separation5Monitoring and auditing4

Impact

Harmful content48Unauthorized actions14Data and privacy leaks19Output manipulation85Availability and reliability8

System type

Models and APIs114LLM applications49AI Agent41web agent10Browser and computer use5Embodied systems and robotics13

Attacker access

Black-box35Gray-box5White-box7

Attack surface / entry point

User input59Web pages and interfaces14Documents and messages7Tool responses4Retrieval corpus27Images, audio and video90Training data10Weights and gradients3Distribution channels3

Layer

Prompt layer66Model layer43Application layer75Training and data layer19

System component

MCP and skills83Memory80RAG51Guardrails and judges172Monitors92Reasoning chains77Permissions and sandboxing93Fine-tuning and open weights65Vision140Audio38Video13Image generation49Embeddings13Multimodal227
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 240papers have no tags yet and are not included in filtered results.
System component：RAGSystem component：VisionSystem component：VideoSystem component：EmbeddingsClear all

Another 240papers do not have category tags yet and are not included in filtered results.

  1. DefensesarXiv 2610.13733

    KVeil: First-Place Solution for the ECCV 2026 LifeGenIP Unlearnable Video Challenge

    Propose KVeil, which adds perturbations to face videos to prevent unauthorized customization from reproducing identity

    arXiv
    2610.13733
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    White-box, Black-box
    Evaluated models
    Wan2.2-TI2V-5B, HunyuanVideo-1.5, Kling 3.0 and others4

    AbstractKVeil adapts LETHE's reference-reading disruption to the challenge's two stages, ranking first on the leaderboard in both Final columns.

    Paper analysisRead analysis
  4. 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
  6. DefensesarXiv 2610.09941

    OrthoPurify 用一步正交投影净化被后门污染的视觉语言模型

    提出 OrthoPurify，用正交投影清除视觉语言模型适配器中的劫持方向

    arXiv
    2610.09941
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    LLaVA-1.5-7B, Qwen3-VL-8B, LLaVA-1.5-13B and others5
    Abstract一次投影去掉劫持方向。

    OrthoPurify 是部署前的 LVLM 后门净化方法，用一次正交投影去掉被劫持的权重更新方向。

    Paper analysisRead analysis
  9. DefensesarXiv 2610.09703

    研究揭示 Token-Pruning 对 VLM 安全性的影响并提出 SAP 缓解机制

    提出 Safety-Aware Pruning，缓解视觉 token 剪枝放大的多模态越狱

    arXiv
    2610.09703
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    LLaVA-1.5-7B
    Abstract剪枝去掉背景 token 后注意力塌向恶意前景。

    这篇工作评估视觉 token 剪枝如何改变 VLM 在多模态越狱下的行为，并给出推理时防御 SAP。

    Paper analysisRead analysis
  10. Analysis and theoryarXiv 2610.09517

    审计发现 CLIP 安全分数主要反映画面与提示词场景的相似度

    开环审计冻结视觉语言安全分，发现其反映场景模板而非危险感知

    arXiv
    2610.09517
    Published
    Layer
    Model layer
    Setting
    Embodied systems and robotics
    Evaluated models
    CLIP ViT-B/32, CLIP ViT-L/14, Qwen2-VL-7B
    Abstract论文审计了冻结 VLM 安全评分机制，指出其反映场景模板而非危险感知，为安全强化学习评估提供了严谨基准。

    本文是一项针对安全强化学习中冻结视觉语言模型（VLM）安全评分机制的实证审计研究。

    Paper analysisRead analysis
  11. DefensesarXiv 2610.09469

    Secure-CUA: Constraining the Impact of Untrusted Content in Computer-Use Agents

    提出 Secure-CUA，用动作事务约束计算机使用智能体的不可信影响

    arXiv
    2610.09469
    Published
    Layer
    Application layer
    Setting
    web agent
    Evaluated models
    Claude Opus 5, Gemini 3.8 Flash, GPT-5.6-Sol
    AbstractSecure-CUA 用每步事务约束不可信影响，良性 TSR 接近 Vanilla-CUA。

    Secure-CUA 要限制图形界面里不可信内容对 CUA 决策和 GUI 执行的影响，同时保留合法任务所需的不可信信息。

    Paper analysisRead analysis
  12. AttacksarXiv 2610.09240

    WebMirage hijacks visual web agents' browser operations with adversarial images

    提出 WebMirage，用局部对抗图像劫持视觉网页 Agent 的浏览器操作

    arXiv
    2610.09240
    Published
    Layer
    Application layer
    Setting
    web agent
    Evaluated models
    LLaVA-v1.5-13B, LLaVA-v1.6-34B, MiniCPM-o-8B and others6
    Abstract论文提出了针对网络智能体的端到端视觉红队框架 WebMirage，揭示了从视觉定位到浏览器执行的安全脆弱性。

    WebMirage 是一个面向视觉定位网络智能体的流水线感知端到端红队攻击框架。

    Paper analysisRead analysis
  13. AttacksarXiv 2610.09027

    P-VMI：对抗图像经 KV cache 持续影响多轮对话

    提出持久视觉记忆注入，使对抗图像经缓存在触发时操控输出

    arXiv
    2610.09027
    Published
    Layer
    Infrastructure layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-8B, Qwen3-VL-8B-Instruct, Gemma-4-12B

    Abstract论文提出 P-VMI，证实对抗输入被掩蔽后仍可经 KV 缓存持久控制模型，强调了保留状态的安全风险。

    Paper analysisRead analysis
  16. Evaluation and benchmarksarXiv 2610.07939

    CCDF dataset: Grok Imagine, Veo 3.1, Sora 2 forge surveillance footage, ten deepfake detectors basically cannot recognize it

    提出监控深伪基准 CCDF，在统一编码后评测检测器对伪造犯罪事故录像的识别

    arXiv
    2610.07939
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    C2P-CLIP, ForgeLens, D3-Im and others10

    Abstract提出消除格式捷径的监控深度伪造基准 CCDF，揭示现有检测器难以识别商用生成视频。

    Paper analysisRead analysis
  17. DefensesarXiv 2610.07774

    研究者提出读取路由 logits 的 MoE 视觉语言模型多模态安全检测器

    AI summary研究者提出一种轻量级路由 logits 安全检测器，在提示词预填充阶段读取 MoE 视觉语言模型的专家路由信号，在生成前判断多模态请求是否安全，且不修改模型参数或专家路由。

    arXiv
    2610.07774
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-VL-30B-A3B-Instruct, Kimi-VL-A3B-Instruct
    Abstract读出而非操纵 router logits，可在不改模型的情况下降低两模型的组合式安全错误。

    作者在两个指令微调 MoE VLM 上比较改模型与读路由。组合式风险按 HoliSafe 分成五类：图文各自可安全，合在一起才不安全，或不安全内容只在图像里。

    Paper analysisRead analysis
  18. DefensesarXiv 2610.07276

    SAFESHIELD：小语言模型部署时安全的决策组织框架

    提出 SAFESHIELD，组织小语言模型部署时的安全决策

    arXiv
    2610.07276
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    Qwen3-4B-Instruct
    AbstractSAFESHIELD 用可审计决策轨迹组织 SLM 部署时的四项安全决策，并报告协调依赖影响端到端结果。

    SAFESHIELD 是面向小语言模型的部署时安全框架，把运行时安全从单项审核改写成可检查的决策过程。

    Paper analysisRead analysis
  20. OtherarXiv 2610.06452

    研究者提出多模态安全评测应衡量可控性而非仅做分类

    主张多模态安全评测应衡量可控性，并用 SAE 建立四轴画像

    arXiv
    2610.06452
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    LlavaGuard-v1.2-7B-OV, Qwen3.5-9B
    Abstract四轴画像把读出与选择性控制分开。

    作者主张多模态安全评测在行为分类之外，单独报告可控性画像，而不是把检测准确当成控制证据。

    Paper analysisRead analysis
Load more