Skip to content

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

Oct 13 · Tuesday

All papers

Found 6 papers
Filter6

Attack type

Prompt injection3

Attack technique

Automated search9Encoding and obfuscation4Reasoning-chain manipulation6Defense-targeted3Dormant triggers26

System type

AI Agent4

Layer

Prompt layer9Model layer1Application layer5

System component

Memory1Monitors1Reasoning chains6
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 203papers have no tags yet and are not included in filtered results.
Attack technique：Reasoning-chain manipulationLayer：Model layerLayer：Application layerSystem component：MemorySystem component：MonitorsSystem component：Reasoning chainsClear all

Another 203papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2609.27542

    Researchers use control token injection to delete gpt-oss-20b's reasoning chain and bypass chain-of-thought monitoring

    提出控制标记注入，清空工具 Agent 的思维链以绕过推理监控

    arXiv
    2609.27542
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    openai/gpt-oss-20b, google/gemma-4-26B-A4B-it, deepseek-ai/DeepSeek-R1-Distill-Qwen-7B and others4

    Abstract论文阐明控制标记注入可抑制 CoT 使监控失效并促成拒绝绕过，且工具执行受解析器宽容度影响。

    Paper analysisRead analysis
  2. Risky behaviorarXiv 2609.08186

    论文揭示大推理模型的对齐崩溃并提出 Reasoning Trap 越狱范式

    揭示推理加深引发的对齐坍塌，用 Reasoning Trap 放大越狱

    arXiv
    2609.08186
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-8B, Qwen3-14B, Llama2-7B and others8
    Abstract更深推理提升题目准确率，同时提高扰动下的相对损失。

    作者研究扩展推理是否以对齐稳健性为代价，提出 ALR、RT 与 RRA。。

    Paper analysisRead analysis
  5. AttacksarXiv 2604.05549

    Stop obsessing over prompts: reasoning hijacking and constraint tightening for red-teaming LLM agents

    提出 JailAgent，不改用户提示，经推理劫持与记忆投毒操纵 Agent 动作

    arXiv
    2604.05549
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-3.5-turbo, GPT-4o, GPT-5 and others7
    AbstractJailAgent 不改提示词，用三阶段操纵推理与检索。

    JailAgent 是不修改用户原始提示词的 LLM 智能体红队框架。

    Paper analysisRead analysis
  6. AttacksarXiv 2603.12277

    Prompt injection is role confusion

    提出 CoT Forgery，经用户消息或工具输出注入伪造推理

    arXiv
    2603.12277
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Attacker
    Black-box
    Evaluated models
    gpt-oss-20b, gpt-oss-120b, o4-mini and others11
    Abstract提示注入被作者解释为角色感知失败：听起来像某角色，就在表示里成为该角色。

    作者把提示注入解释成角色感知失败：标签划出的权限没有进入模型的内部几何。

    Paper analysisRead analysis
You have reached the end