Skip to content

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

Oct 12 · Monday

All papers

Found 79 papers
Filter5

Attack type

Jailbreaks221Prompt injection79Poisoning and backdoors86Model tampering35Extraction and theft65Detection evasion53Denial of service5Malicious use20

Attack technique

Gradient and representation optimization4Automated search6Encoding and obfuscation5Role-play and persuasion3Cross-modal payloads5Defense-targeted4Dormant triggers8Selection manipulation3

Model risks

Deception and scheming4Misalignment and value divergence3Dangerous agent actions3

Defense type

Detection and filtering13Instruction–data separation7Permissions and approval11Monitoring and auditing8

Impact

Harmful content3Unauthorized actions31Data and privacy leaks10Output manipulation25Oversight failure4

System type

Models and APIs16LLM applications24AI Agent209coding agent33science agent3web agent24Browser and computer use8Multi-agent systems24

Attack surface / entry point

User input28Web pages and interfaces11Documents and messages13Tool responses17Tool and skill descriptions18Code repositories and dependencies13Retrieval corpus5Memory writes9Inter-agent messages17Images, audio and video11Distribution channels13

Layer

Prompt layer11Model layer4Application layer62

System component

MCP and skills17Memory10Guardrails and judges11Monitors12Reasoning chains3Permissions and sandboxing14Vision10Multimodal11
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 254papers have no tags yet and are not included in filtered results.
Attack type：Prompt injectionSystem type：Models and APIsSystem type：coding agentSystem type：Browser and computer useSystem type：Multi-agent systemsClear all

Another 254papers do not have category tags yet and are not included in filtered results.

  1. InterpretabilityarXiv 2610.12361

    Study: legal reasoning models citing legal provisions does not equal relying on legal provisions.

    Uses premise permutation and hidden-layer tracing to audit whether legal chain-of-thought truly relies on the cited legal provisions.

    arXiv
    2610.12361
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-8B, Qwen3-14B, Llama-3.1-8B and others7
    AbstractUsing counterfactual substitution of authorities and internal commitment tracing, the authors find that legal CoT rulings depend on authorities only to a limited extent and are vulnerable to prompt injection manipulation.

    This study conducts an empirical audit of the causal truthfulness of legal citations in chain-of-thought reasoning of large language models in the legal domain.

    Paper analysisRead analysis
  3. AttacksarXiv 2610.09264

    研究者提出 PackHallu：经规则文件提示注入对编码 Agent 发起包幻觉攻击

    提出 PackHallu，经规则文件提示注入诱导编码 Agent 换用攻击者指定包

    arXiv
    2610.09264
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Qwen2.5-Coder-7B, Qwen3-Coder-30B, Devstral-Small-2-24B and others19
    AbstractPackHallu 诱导规则文件换包。

    PackHallu 研究一种针对编程智能体规则文件的提示注入：让智能体在本应使用合法依赖时，改导入攻击者指定并控制的包。

    Paper analysisRead analysis
  4. 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
  6. DefensesarXiv 2610.07657

    DEFER1 多智能体防御研究报告受控测试中的攻击成功率下降

    实现 DEFER，以确定性检查优先拦截多智能体流水线的内容攻击

    arXiv
    2610.07657
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Qwen3-235B-A22B-Instruct-2507, gpt-oss-120b, Llama-4-Scout-17B-16E-Instruct and others5
    AbstractDEFER 把确定性检查放在 judge 之前。

    DEFER 用来测量多智能体授权流水线里，确定性规则在何处结束、LLM judge 从何处开始。

    Paper analysisRead analysis
  8. Analysis and theoryarXiv 2610.06001

    AgentSpy: Observing AI agent behavior at the system call layer and detecting skill injection attacks

    提出 AgentSpy，在系统调用层观测 Agent 行为并检测技能注入

    arXiv
    2610.06001
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    gpt-5.6-terra, deepseek-v4-flash, glm-5.3-flash and others4
    Abstract提出了系统级智能体监控框架 AgentSpy，揭示出通过测试背后的非预期行为与注入风险。

    本研究提出了面向 AI 智能体的系统级外部监控与行为分析框架 AgentSpy。

    Paper analysisRead analysis
  9. DefensesarXiv 2610.05640

    Research reveals that the CaMeL defense fails in multi-agent systems and proposes multi-CaMeL

    提出 multi-CaMeL，在智能体调用边界隔离可信指令与非可信数据

    arXiv
    2610.05640
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Claude Fable 5, GPT-5.5, GPT-4.1 and others6

    Abstract论文针对多智能体中单体防御不组合的边界清洗问题，提出 multi-CaMeL 协议彻底阻断了 MultiAgentDojo 上的注入攻击。

    Paper analysisRead analysis
  10. DefensesarXiv 2610.05163

    Researchers construct segmented prompt injection on Claude Code and Codex and release a boundary action audit benchmark

    提出 PAA 方法，在边界动作前审计长流程智能体的分阶段提示注入

    arXiv
    2610.05163
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude Code (v2.1.172, Claude Sonnet 5), Codex CLI (v0.144.0, GPT-5.5)
    Abstract论文针对长流程注入提出边界动作审计与 PAA，在保持低误阻的同时实现高拦截率。

    本研究针对长流程智能体在执行任务时面临的分阶段间接提示注入威胁，提出了在动作生效前进行拦截的边界动作审计框架与路径对齐归因方法。

    Paper analysisRead analysis
  12. DefensesarXiv 2610.04412

    论文研究 LLM 数字孪生参与增强民主时的提示注入脆弱性

    提出检测、结构化与 GSPO 流程，降低共识生成的提示注入错位

    arXiv
    2610.04412
    Published
    Layer
    Application layer
    Setting
    LLM applications, Multi-agent systems
    Evaluated models
    LLaMA-2 7B, LLaMA-3 8B, GPT 3.5 Turbo and others14
    Abstract三章依次做个体偏好预测、政党知识图谱审议，以及共识生成对提示注入的脆弱性与一条防御流程。

    这是一篇把 LLM 数字孪生放进增强民主的博士论文，分个体表示、政党聚合和共识生成的提示注入三章。

    Paper analysisRead analysis
  14. DefensesarXiv 2610.03073

    SecJev：为 System One 决策模型引入安全专业知识

    提出 SecJev，用场景加权训练把决策模型适配为安全类型化判断

    arXiv
    2610.03073
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    SecJev-0.8B, SecJev-2B, SecJev-4B and others4
    AbstractSecJev 用共享打分器做安全决策，0.8B 的 task-macro 超过 Kev-9B。

    SecJev 是 0.8B 到 9B 的安全决策模型族，作者称据其所知是首个面向安全的 Jev-like 家族。

    Paper analysisRead analysis
  15. DefensesarXiv 2610.02413

    研究：分类后缀可提升激活探针的分布外恶意输入检测

    用分类后缀提升激活探针对分布外恶意输入的检测

    arXiv
    2610.02413
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Llama-3.1-8B, Qwen3.5-9B, Gemma-4-12B
    Abstract作者称排序增益来自分类格式，低 FPR 精度来自点名准则。

    用户轮后的分类后缀，被作者用作激活探针的分布外检测干预：排序看格式，低 FPR 精度看是否点名恶意/良性。

    Paper analysisRead analysis
  16. DefensesarXiv 2610.03089

    COBRA：为计算机使用 Agent 防御分支引导攻击的双 LLM 架构

    提出 COBRA 双 LLM 架构，防御计算机使用 Agent 的分支转向攻击

    arXiv
    2610.03089
    Published
    Layer
    Application layer
    Setting
    Browser and computer use
    Evaluated models
    gpt-5, UI-TARS-1.5-7B, claude-sonnet-5 and others6

    Abstract提出分支转向防御架构 COBRA，通过事前能力约束与边界代理消除分支转向攻击并兼顾效用。

    Paper analysisRead analysis
  18. AttacksarXiv 2610.00392

    A2A-TIBA：针对 A2A 协议 LLM 智能体的间接提示注入攻击与三层同构攻防模型

    提出 A2A-TIBA，经 A2A 对等任务间接注入并植入常驻回调

    arXiv
    2610.00392
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    MiniMax-M3, deepseek-v4-pro, glm-5.3 and others8
    AbstractA2A-TIBA 一次植入后绕过智能体，GDA 用四类结果定位防线，ELA-ITL 把信封标注列为第三层。

    A2A-TIBA 把间接提示注入和绕过放进 A2A：gray-box 攻击者已能投递任务，诱导目标部署常驻回调程序后，下令和外传不再经过智能体。

    Paper analysisRead analysis
  19. 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
  20. DefensesarXiv 2610.00371

    FlowReview：以授权配对评测控制多智能体组合信息流

    提出 FlowReview，用授权配对与确定性提交门控控制多智能体组合信息流

    arXiv
    2610.00371
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Sonnet 4.5, Haiku 4.5, Qwen3-32B and others5
    Abstract配对标准要求拒绝禁止使用并完成授权使用，FlowReview 把三项审查接到执行。

    这篇工作把多智能体安全写成一件必须同时成立的事：拦住组合信息所启用的禁止使用，并完成同一场景里任务所需的授权使用。

    Paper analysisRead analysis
Load more