Skip to content

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

Oct 12 · Monday

All papers

Found 40 papers
Filter3

Attack type

Prompt injection28Poisoning and backdoors19

Attack technique

Automated search3Dormant triggers5

Model risks

Dangerous agent actions3

Defense type

Model hardening3

Impact

Unauthorized actions15Data and privacy leaks9Output manipulation11

System type

Models and APIs6AI Agent32coding agent6web agent3Multi-agent systems5

Attack surface / entry point

User input98Web pages and interfaces19Documents and messages25Tool responses24Tool and skill descriptions13Code repositories and dependencies7Retrieval corpus8Memory writes11Inter-agent messages8Images, audio and video23Training data6Weights and gradients3Distribution channels8

Layer

Application layer34Training and data layer5

System component

MCP and skills4Memory14Fine-tuning and open weights3Vision4Multimodal5
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 248papers have no tags yet and are not included in filtered results.
Attack surface / entry point：Documents and messagesAttack surface / entry point：Memory writesAttack surface / entry point：Training dataClear all

Another 248papers do not have category tags yet and are not included in filtered results.

  2. Evaluation and benchmarksarXiv 2610.04589

    StegoMemory red team testing: agent persistent memory can be used as a cross-session covert channel

    提出 StegoMemory，评测智能体持久记忆作为跨会话隐写信道

    arXiv
    2610.04589
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Llama 3.1 405B, Claude 3.5 Sonnet, GPT-4o and others13

    Abstract研究揭示智能体持久记忆可作为隐蔽信道跨会话传输机密，写入持久化是主要防御瓶颈，呼吁建立记忆完整性防御。

    Paper analysisRead analysis
  3. 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
  4. Evaluation and benchmarksarXiv 2609.39146

    MADBench：多智能体辩论安全性评测基准发布

    提出 MADBench，评测多智能体辩论是吸收还是放大对抗影响

    arXiv
    2609.39146
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    gpt-4o, GPT-5, GPT-6 Astra
    AbstractMADBench 显示辩论能吸收部分准确率攻击，同时放大未授权读写。

    MADBench 是一个评测多智能体辩论安全性的基准，用来看攻击是被辩论吸收还是被放大。

    Paper analysisRead analysis
  5. Evaluation and benchmarksarXiv 2609.36817

    pikit：面向间接提示注入研究与评测的可组合工具包

    构建可组合工具包 pikit，评测编码智能体的间接提示注入

    arXiv
    2609.36817
    Published
    Layer
    Application layer
    Setting
    coding agent
    Abstractpikit 提供组合式间接注入评测。

    pikit 是面向间接提示注入研究的组合式工具包，主要贡献是组织攻击、投递通道、防御和判定流程，而非提出新的攻击优化算法。

    Paper analysisRead analysis
  8. Evaluation and benchmarksarXiv 2609.32691

    研究审计 Agent 安全评测缺陷：工具调用中介在间接提示注入下的有效评测框架

    提出卡点评测框架以审计智能体间接提示注入评测的测量缺陷

    arXiv
    2609.32691
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    gpt-5.6-terra, llama3.1:8b, gemma4:12b and others4

    Abstract论文揭示了 IPI 评测中导致指标虚高的系统性缺陷，通过构建有效评测框架纠正了多项安全与模型能力结论。

    Paper analysisRead analysis
  9. Evaluation and benchmarksarXiv 2609.32616

    Cave-Bench：虚假指控下 LLM 智能体最高 60.06% 的运行会破坏正确工作

    提出 CAVE-BENCH，评测智能体在虚假指责下破坏已完成的正确工作

    arXiv
    2609.32616
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude-Sonnet-5, Claude-Opus-5, GPT-5.6-Sol and others14

    Abstract论文提出了评估智能体在虚假指责下破坏已有正确成果的 CAVE-BENCH，发现强模型易推翻已知证据，提出了门控缓解方案。

    Paper analysisRead analysis
  11. 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
  13. Evaluation and benchmarksarXiv 2609.24662

    DUMA-Bench：评估 LLM Agent 安全的双控多智能体基准

    提出双控多智能体基准 DUMA-Bench，测量对抗环境下的策略违反

    arXiv
    2609.24662
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    GPT-5, GPT-5-mini, GPT-5-nano and others14
    AbstractDUMA-Bench 用双控交互评测智能体安全。

    DUMA-Bench 把 τ²-bench 的双控交互做成智能体安全评测，用来看策略违反如何随用户、模型和环境一起出现。

    Paper analysisRead analysis
  16. 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
  17. Evaluation and benchmarksarXiv 2609.10952

    监督学习中的数据投毒攻击实证评估：标签翻转与后门投毒在 MNIST 和 Fashion-MNIST 上的对比

    比较标签翻转与后门投毒对图像分类器的影响

    arXiv
    2609.10952
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Logistic Regression, Linear SVM, Random Forest
    Abstract标签翻转更能拉低线性模型。

    Khan 与 Abusaqer 在两个图像分类基准上比较了两种训练时投毒。

    Paper analysisRead analysis
  18. Evaluation and benchmarksarXiv 2609.09404

    MMPIBench：多模态提示注入对六种 Agent 框架的跨框架评测

    用 MMPIBench 评测多模态提示注入对 Agent 框架的影响

    arXiv
    2609.09404
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Claude Opus 4.8, GPT-5.4, Gemini 3.1 Pro and others6

    Abstract论文提出 MMPIBench 评测多模态提示注入对智能体的影响，发现模型主导行为、尝试远超完成，音频通道缺乏防护。

    Paper analysisRead analysis
  19. Evaluation and benchmarksarXiv 2608.28411

    LongPIBench：面向长上下文提示注入的评测基准

    提出 LongPIBench，评测长文档场景下的提示注入攻击与防御

    arXiv
    2608.28411
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    Llama-3.1-8B-Instruct, Llama-3.2-3B-Instruct, Apertus-8B-Instruct and others11

    Abstract论文构建长文本提示注入基准 LongPIBench，揭示启发式与优化攻击在长文档中的高有效性及现有防御的大幅失效。

    Paper analysisRead analysis
Load more