Skip to content

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

Oct 13 · Tuesday

All papers

Found 12 papers
Filter6

Attack type

Prompt injection6Poisoning and backdoors7

Attack technique

Encoding and obfuscation3Dormant triggers5

Impact

Output manipulation6

System type

LLM applications40AI Agent12coding agent2web agent3Browser and computer use1

Attack surface / entry point

User input5Retrieval corpus5Weights and gradients3Distribution channels3

Layer

Application layer12Training and data layer4

System component

MCP and skills82Memory76RAG9Guardrails and judges46Monitors58Reasoning chains11Permissions and sandboxing90Fine-tuning and open weights4Vision21Audio3Multimodal23
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 227papers have no tags yet and are not included in filtered results.
System type：AI AgentSystem type：coding agentSystem type：Browser and computer useLayer：Application layerSystem component：RAGSystem component：Fine-tuning and open weightsClear all

Another 227papers do not have category tags yet and are not included in filtered results.

  1. 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
  2. Evaluation and benchmarksarXiv 2610.05586

    AgentDoxx：联网搜索 Agent 对匿名访谈文本的再识别评测

    构建 AGENTDOXX 基准，评测联网搜索 Agent 对匿名访谈的再识别

    arXiv
    2610.05586
    Published
    Layer
    Application layer
    Setting
    web agent
    Evaluated models
    GPT-5.6 Terra, GPT-5 Mini, GPT-4.1 Mini and others10

    Abstract论文提出了已知身份的重识别基准 AGENTDOXX，揭示搜索与记忆双重风险及现有防御在检索轨迹中的泄露漏洞。

    Paper analysisRead analysis
  3. 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
  4. AttacksarXiv 2609.37468

    论文提出针对 Agentic RAG 的检索器后门攻击与 DOU 隐藏方法

    提出检索器后门攻击操纵智能体搜索，并用 DOU 逃避检测

    arXiv
    2609.37468
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    E5, Contriever-MSMARCO

    Abstract论文展示了通过恶意检索器操纵智能体多轮搜索并利用诱饵重塑表征隐蔽后门的可行性与系统风险。

    Paper analysisRead analysis
  5. OtherarXiv 2609.16694

    面向安全 AI 渗透测试智能体：安全威胁、护栏与架构视角

    综述自主渗透智能体的继承攻击、架构攻击及护栏缺口

    arXiv
    2609.16694
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Abstract用双轴分类连接渗透智能体的架构、攻击与护栏，并认为现有防御仍按组件分开。

    这是一篇关于自主 AI 渗透测试智能体的安全综述：用架构和生命周期两轴整理威胁与护栏，不做新的攻击实验。

    Paper analysisRead analysis
  6. AttacksarXiv 2608.06862

    SynChain：诱导计算机使用 Agent 自建攻击链并跨任务持久化

    提出 SYNCHAIN，诱导计算机使用 Agent 自建跨任务持久攻击链

    arXiv
    2608.06862
    Published
    Layer
    Application layer
    Setting
    Browser and computer use, coding agent
    Evaluated models
    Qwen3.5-9B, Llama-3.1-8B, Ministral-3-8B

    Abstract论文揭示了 CUA 自合成工件带来的内部供应链威胁，验证了潜伏载荷的跨任务传播能力与防御局限。

    Paper analysisRead analysis
  7. AttacksarXiv 2608.04565

    Researchers propose Breadcrumbing long-range attack that can hijack the evidence chain of DeepResearch-style search agents

    提出权威链劫持与轨迹引导策略演化，劫持搜索 Agent 的证据链

    arXiv
    2608.04565
    Published
    Layer
    Application layer
    Setting
    web agent
    Evaluated models
    Qwen3.5-9B, Gemma4-31B, DeepResearch and others7

    Abstract论文揭示了多步搜索智能体在受限中间人操纵下的脆弱性，提出 ACH 策略与 TGSE 演化方法并完成系统验证。

    Paper analysisRead analysis
  9. AttacksarXiv 2607.00422

    KidnapRAG：黑盒投毒攻击劫持 Agentic RAG 的推理链

    提出 KidnapRAG，用投毒文档黑盒劫持 Agentic RAG 的推理链

    arXiv
    2607.00422
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Qwen2.5-32B-Instruct, Llama-3.3-70B-Instruct, QwQ-32B and others5
    Abstract顺序投毒可把多步推理拖向指定应答，作者称关键在持续引导而非单次注入。

    KidnapRAG 是针对 Agentic RAG 的 black-box 语料投毒：攻击者不能访问模型、提示词、检索器或私有轨迹，只能发布可能被检索的文档，并利用公开界面上的推理与搜索查询。

    Paper analysisRead analysis
  10. AttacksarXiv 2603.17239

    LAAF: A Systematic Red-Teaming Method and Automated Attack Framework for LPCI Vulnerabilities in Agentic LLM Systems

    Proposes LAAF, which automatically attacks LPCI vulnerabilities in agent systems in stages

    arXiv
    2603.17239
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    gemini-2.0-flash-001, claude-3-haiku-20240307, meta-llama/llama-3.1-70b-instruct and others5
    AbstractLAAF does LPCI red teaming with taxonomy plus stage seeding.

    The authors state that LAAF is an automated red-teaming framework built specifically for LPCI in agentic LLMs. The issue to address is that encoded, delayed, and conditionally triggered payloads can persist in memory or RAG and re-enter across sessions, whereas Garak and PyRIT do not use the previous round's breakthrough results to seed the next stage across the six stages.

    Paper analysisRead analysis
  12. AttacksarXiv 2507.10457

    论文提出 LPCI：针对 Agent 逻辑层与持久记忆的提示控制注入攻击

    提出 LPCI，把延迟条件载荷写入记忆与检索库以跨会话触发

    arXiv
    2507.10457
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    ChatGPT, Claude, LLaMA3 and others8
    Abstract作者定义 LPCI。

    作者把LPCI定义为针对智能体逻辑执行、记忆留存和工具集成的跨阶段漏洞类，而不是单次输入上的提示改写。

    Paper analysisRead analysis
You have reached the end