Skip to content

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

Oct 13 · Tuesday

All papers

Found 45 papers
Filter6

Attack type

Prompt injection29Poisoning and backdoors25

Attack technique

Gradient and representation optimization5Cross-modal payloads5Dormant triggers9

Defense type

Model hardening4Detection and filtering5Instruction–data separation5Permissions and approval4Monitoring and auditing5

Impact

Harmful content4Unauthorized actions17Output manipulation17

System type

Models and APIs13AI Agent30web agent7Browser and computer use4Multi-agent systems5

Attack surface / entry point

User input112Web pages and interfaces24Documents and messages26Tool responses32Tool and skill descriptions13Code repositories and dependencies4Retrieval corpus6Memory writes40Inter-agent messages6Images, audio and video120Training data15Weights and gradients11Distribution channels14

Layer

Prompt layer4Application layer30Training and data layer10

System component

MCP and skills13Memory14RAG5Guardrails and judges15Monitors8Reasoning chains7Permissions and sandboxing21Fine-tuning and open weights37Vision18Video3Multimodal24
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 215papers have no tags yet and are not included in filtered results.
Attack surface / entry point：Web pages and interfacesAttack surface / entry point：Inter-agent messagesAttack surface / entry point：Training dataSystem component：MemorySystem component：MonitorsSystem component：MultimodalClear all

Another 215papers 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
  3. 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
  4. DefensesarXiv 2610.09793

    用时序逻辑监控工具 Agent

    提出 MFOTL 离线时序验证以监控工具 Agent 的危险动作链

    arXiv
    2610.09793
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Attacker
    Black-box
    Abstract形式化验证可有效识别多步攻击，但其有效性高度依赖轨迹的可信历史记录。

    这篇论文对工具调用大语言模型智能体的安全监控进行了形式化时序验证研究。

    Paper analysisRead analysis
  5. 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
  6. 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
  7. Evaluation and benchmarksarXiv 2610.03153

    EvoRiskBench: An evolutionary benchmark for runtime security risks of workspace agents

    构建 EvoRiskBench 评测工作区 Agent 的运行时安全风险

    arXiv
    2610.03153
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5.6 Sol, DeepSeek-V4-Pro-0813, Claude Opus 5

    Abstract论文构建了工作空间智能体运行时风险基准 EvoRiskBench，发现高执行力智能体在 IPI 下面临较高安全风险。

    Paper analysisRead analysis
  10. Evaluation and benchmarksarXiv 2610.06898

    DIBench：面向移动 GUI 智能体多候选选择决策完整性的安全基准

    用 DIBench 评测移动 GUI 智能体被 UI 注入后的决策完整性

    arXiv
    2610.06898
    Published
    Layer
    Application layer
    Setting
    Browser and computer use
    Evaluated models
    Qwen2.5-VL(32B), Qwen3-VL(32B), GUI-Owl(32B) and others7

    Abstract论文提出移动智能体决策完整性基准 DIBench，发现欺诈 UI 注入诱导目标选择并推高完成率，通用防御收益不稳定。

    Paper analysisRead analysis
  11. DefensesarXiv 2609.35659

    Tracekit：面向自主编码 Agent 的防篡改意图-推理-动作审计系统

    提出 Tracekit，审计编码 Agent 的意图、推理与动作并检测账本篡改

    arXiv
    2609.35659
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude Code (claude -p, version 2.1), 独立评审（同一 CLI，无工具）
    AbstractTracekit 把意图、自述和动作写入可锚定的哈希链。

    Tracekit 是无第三方依赖的编码智能体审计层：把人类请求、模型自述和已执行动作并排写入可外部锚定的哈希链，执行前用正则门，事后用规则和独立模型互查。

    Paper analysisRead analysis
  13. Evaluation and benchmarksarXiv 2609.33102

    ORBIT：基于 UK AISI Inspect 的多智能体安全评测框架

    提出 ORBIT 框架，评测多智能体防御对合谋等威胁的效果

    arXiv
    2609.33102
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    GPT-5.4, Claude Sonnet 4.6, Qwen3-235B and others5

    AbstractORBIT 用六层配置比较多智能体防御，逐动作监控不迁移到合谋。

    Paper analysisRead analysis
  14. Evaluation and benchmarksarXiv 2609.28915

    研究提出 ACE 语料库，评估内核级证据对 Agent 安全检测的有效性

    构建 ACE 语料，评测内核级证据对 Agent 攻击检测的有效性

    arXiv
    2609.28915
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    AdaBoost, XGBoost, TabPFN and others10
    AbstractACE 用配对的内核与应用层证据表明，syscall 痕迹可判别智能体攻击，两层组合通常更好。

    ACE 是一份把智能体会话的内核 syscall 痕迹与应用层清单、转录配对起来的检测语料和基准。。

    Paper analysisRead analysis
  17. Evaluation and benchmarksarXiv 2609.17320

    Emergence World：对长时程多智能体系统开展对抗压力测试

    提出 Emergence World，对长时程多智能体系统开展对抗压力测试

    arXiv
    2609.17320
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Claude Opus 4.8, DeepSeek v4 Pro, Gemini 3.5 Flash and others9

    Abstract作者通过长期平行世界实验指出模型对齐不具备可组合性，单模型群体会产生社会奉承、协同退缩及语言不透明等失效。

    Paper analysisRead analysis
  18. AttacksarXiv 2609.07048

    FreqDoor：面向视觉语言模型的频域后门攻击

    提出 FreqDoor，在频域混合振幅并对生成式 VLM 做训练时视觉后门

    arXiv
    2609.07048
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    BLIP-2 (opt-2.7b), InstructBLIP (flan-t5-xl), LLaVA (interleave-qwen)
    AbstractFreqDoor 用频域振幅混合给三款 VLM 植入仅视觉的训练时后门。

    FreqDoor 是针对生成式 VLM 的仅视觉、训练时后门：在频域混合触发源振幅、保留干净图像相位，使触发在空间上分散。

    Paper analysisRead analysis
  19. DefensesarXiv 2609.06346

    面向后门攻击下持续学习的鲁棒动态扩展：净化与选择性恢复

    提出持续学习下的净化与选择性恢复框架，抵御逐任务训练投毒后门

    arXiv
    2609.06346
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    ViT-B/16
    Abstract作者报告任务内净化与恢复可配合动态专家防御后门，但保证依赖分离条件。

    这是一项面向持续图像分类的后门防御研究：模型只使用当前任务数据，同时需要保留过去知识并避免学习恶意监督。

    Paper analysisRead analysis
Load more