Skip to content

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

Oct 12 · Monday

All papers

Found 184 papers
Filter3

Attack type

Jailbreaks16Prompt injection147Poisoning and backdoors32Detection evasion15Malicious use5

Attack technique

Gradient and representation optimization3Automated search28Multi-turn escalation3Encoding and obfuscation8Role-play and persuasion7Conversation structure manipulation4Reasoning-chain manipulation3Defense-targeted6Dormant triggers13

Model risks

Misalignment and value divergence4Dangerous agent actions7

Defense type

Model hardening8Detection and filtering24Instruction–data separation28Permissions and approval25Monitoring and auditing15Watermarking and provenance11

Impact

Harmful content8Unauthorized actions91Data and privacy leaks20Output manipulation45Availability and reliability3Oversight failure8

System type

Models and APIs14LLM applications21AI Agent149coding agent25web agent7Browser and computer use5Multi-agent systems11

Attacker access

Black-box13White-box4

Attack surface / entry point

User input410Web pages and interfaces69Documents and messages87Tool responses108Tool and skill descriptions63Code repositories and dependencies31Retrieval corpus37Memory writes39Inter-agent messages27Images, audio and video124Training data71Weights and gradients69Distribution channels60Inference services and hardware22

Layer

Prompt layer5Model layer8Application layer170

System component

MCP and skills21Memory25RAG6Guardrails and judges23Monitors16Reasoning chains6Permissions and sandboxing28Vision5Multimodal5
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 261papers have no tags yet and are not included in filtered results.
Attack surface / entry point：Documents and messagesAttack surface / entry point：Tool responsesAttack surface / entry point：Code repositories and dependenciesClear all

Another 261papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2610.12292

    The paper proposes an option channel attack: changing one option label can make an agent guardrail allow a prohibited operation

    Proposing an option channel attack that renames option labels so that agent guardrails allow prohibited operations

    arXiv
    2610.12292
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Attacker
    Black-box
    Evaluated models
    LAYA-TD, LAYA-EN, LAYA-ML and others5

    AbstractThe authors' evaluation indicates that typed models should not serve as final safety arbiters; they are susceptible to off-policy text and option name manipulation, and confidence filtering cannot resist attacks.

    Paper analysisRead analysis
  2. DefensesarXiv 2610.11848

    SemField: training-free semantic watermark

    Proposes the semantic watermark SemField, which still identifies the attribution of generated text after paraphrasing and structural changes.

    arXiv
    2610.11848
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Llama-3.2-1B, Qwen3-1.7B, OPT-1.3B
    AbstractTraining-independent document-level semantic watermark.

    SemField is a zero-bit semantic watermark embedded by the provider at generation time and detected later with a key, used to identify the provider's own model text after paraphrasing and sentence-level structural changes.

    Paper analysisRead analysis
  3. DefensesarXiv 2610.11634

    LTBD：用可学习信任边界分隔符防御提示注入

    提出 LTBD，用可学习分隔符隔离可信指令与不可信数据

    arXiv
    2610.11634
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    Llama3-8B-Instruct, Llama3.1-8B-Instruct, Falcon3-7B-Instruct and others4
    AbstractLTBD 只训练四个分隔符嵌入。

    LTBD 是一种不改基座权重的提示注入防御：只学习用来标出可信指令与不可信数据边界的四个 token 嵌入。

    Paper analysisRead analysis
  6. 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
  7. 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
  9. AttacksarXiv 2610.08951

    ASPIRE：面向 LLM Agent 的开放式提示注入红队引擎

    提出 ASPIRE 行为图搜索，开放式发现 Agent 提示注入漏洞

    arXiv
    2610.08951
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Gemini-3.7-Flash, DeepSeek-V4-Flash, Gemini-2.5-flash
    AbstractASPIRE 把提示注入红队从载荷优化改成行为图引导的开放式漏洞发现。

    ASPIRE 是给搭建方用的智能体提示注入红队引擎，目标是开放式画出可利用的行为路径，而不是在固定任务上打磨一条载荷。

    Paper analysisRead analysis
  10. DefensesarXiv 2610.08668

    SBW：面向 LLM 智能体的语义行为水印方法

    提出 SBW，为 LLM Agent 轨迹做抗改写与伪造的语义行为水印

    arXiv
    2610.08668
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Qwen2.5-3B-Instruct, LLaMA-3.2-3B, Qwen2.5-7B and others5
    AbstractSBW 在语义簇上做溯源水印，改写检测高于精确符号，链式重放仍开放。

    SBW 是免训练的智能体行为水印，把所有者标识放在离散动作选择上，而不是输出 token。

    Paper analysisRead analysis
  12. 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
  14. AttacksarXiv 2610.07510

    PersistBD: attackers can strengthen backdoors before release so that they maintain high attack success rates even after developer SFT and RL.

    提出 PERSISTBD，发布前强化供应链后门使其经良性 SFT 与 RL 后仍外发凭据

    arXiv
    2610.07510
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Qwen2.5-Coder-3B-Instruct, Qwen2.5-Coder-7B-Instruct, Qwen3-Coder-30B-A3B-Instruct
    Abstract总结了论文关于后门在 SFT/RL 中存活的发现、两因子机制、PERSISTBD 方法与核心数字。

    本文研究第三方 LLM 智能体在开发者进行良性监督微调（SFT）和强化学习（RL）后训练时的后门持久性风险。

    Paper analysisRead analysis
  16. Evaluation and benchmarksarXiv 2610.07089

    研究者提出统一滥用监控基准，覆盖分解攻击与提示注入约 6200 条轨迹

    提出危害窗口与区间指标，统一评测分解攻击和提示注入下的智能体滥用监视

    arXiv
    2610.07089
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Gemma 4 26B-A4B, Qwen3.6 27B, Qwen3.5 27B and others12

    Abstract统一智能体滥用监视框架，动作视角兼顾两类威胁但危害定位仍存瓶颈。

    Paper analysisRead analysis
  17. DefensesarXiv 2610.05640

    研究揭示 CaMeL 防护在多智能体系统中失效并提出 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
  18. InterpretabilityarXiv 2610.05295

    研究用因果追踪区分间接提示注入中的可读信号与有效干预

    用因果追踪区分间接提示注入中的角色可读性与有效干预

    arXiv
    2610.05295
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Qwen-2.5-7B, Qwen-2.5-1.5B, Llama-3.1-8B and others8
    Abstract作者称源角色先于强工具选择效应可读，有效编辑还取决于位置、长度和方向来源。

    在权重固定的白盒模型上，源角色可以先被读出。沿相关方向的编辑是否改变行为，要在目标位点上另测。。

    Paper analysisRead analysis
  19. AttacksarXiv 2610.05266

    论文披露主动式智能体的服务方间接提示注入攻击

    提出服务方间接提示注入，把主动智能体的决策协助转向预选目标

    arXiv
    2610.05266
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5, Claude Opus 4.6, Gemini 3 Pro and others7
    Abstract提供者侧间接注入用 T–B–S 把主动协助转向外部目标，并在三环境中提高模拟用户授权。

    外部提供者只控制与自身目标关联的内容，就可以把良性主动个人智能体的协助转向该目标。

    Paper analysisRead analysis
  20. 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
Load more