Skip to content

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

Oct 12 · Monday

All papers

Found 27 papers
Filter5

Attack type

Jailbreaks23

Attack technique

Gradient and representation optimization17Automated search48Multi-turn escalation22Encoding and obfuscation42Role-play and persuasion23Conversation structure manipulation10Reasoning-chain manipulation10Cross-modal payloads26Defense-targeted9

Impact

Harmful content24

System type

Models and APIs27

Attacker access

Black-box14White-box12

Attack surface / entry point

User input27Images, audio and video25Weights and gradients10

Layer

Prompt layer24Model layer3Application layer5

System component

Guardrails and judges4Reasoning chains3Vision3Multimodal3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 255papers have no tags yet and are not included in filtered results.
Attack technique：Gradient and representation optimizationAttack technique：Conversation structure manipulationAttack surface / entry point：User inputLayer：Prompt layerLayer：Model layerClear all

Another 255papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2609.36956

    研究者提出 Controlled Decoding Attacks，通过纯文本接口越狱黑盒 LLM

    提出 BLINDBIAS，通过纯文本续写与前缀控制越狱黑盒模型

    arXiv
    2609.36956
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    GLM-5, Gemini-3.5-Flash, Qwen3-32B and others4
    Abstract作者认为，隐藏数值概率并不足以消除支持重复采样和前缀续写的解码攻击面。

    BLINDBIAS 是一种针对 sample-only 续写接口的 black-box 解码时越狱框架，研究没有目标数值概率时，采样能否支持逐步控制。

    Paper analysisRead analysis
  3. InterpretabilityarXiv 2609.24801

    研究用 XAI 归因分析 Prompt Guard 2，同义词替换可翻转其提示注入判定

    用词级归因指导同义词替换，翻转提示注入检测并验证越狱

    arXiv
    2609.24801
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Prompt Guard 2 (86M), Llama 3.1 8B

    Abstract作者称检测靠分散的词法线索。

    Paper analysisRead analysis
  5. AttacksarXiv 2609.02172

    BOSS：以广度优先后缀搜索改进 GCG 越狱优化

    提出广度优先后缀搜索 BOSS，改进 GCG 越狱优化

    arXiv
    2609.02172
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    White-box
    Evaluated models
    Llama-2-7B-Chat, Yi-1.5-9B-Chat, Qwen2-7B-Instruct and others8
    AbstractBOSS 把 GCG 类优化从单条深搜索改为多条短轨迹加源侧诊断。

    BOSS 是插在 GCG 类离散 suffix 优化上的搜索框架：局部梯度更新不变，变的是固定预算花在深度还是广度。

    Paper analysisRead analysis
  7. AttacksarXiv 2608.09867

    Researchers steal proprietary LLM inference chains via cross-model compatibility of encrypted inference blocks

    利用加密思维块的跨模型重放，诱导弱模型逐字转写专有推理链

    arXiv
    2608.09867
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Claude Opus 4.8, Claude Sonnet 4.6, Claude Haiku 4.5 and others13

    Abstract论文揭示了客户端加密思维块跨模型互换漏洞，提出了低成本提取方法并给出防御方案。

    Paper analysisRead analysis
  8. AttacksarXiv 2608.07430

    研究者提出 SN-Guided Diffusion，利用扩散大模型安全神经元实现离线越狱

    提出 SN-Guided Diffusion，利用安全神经元实现离线越狱

    arXiv
    2608.07430
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box, White-box
    Evaluated models
    Qwen2.5-7B-Instruct, LLaDA-8B-Instruct, Dream-Instruct-7B and others15
    Abstract论文揭示了扩散语言模型对自回归安全神经元的继承性，并提出纯离线黑盒越狱框架 SN-Guided Diffusion。

    论文系统评估了扩散大语言模型（DLLM）在安全对齐中的双重角色——既作为白盒攻击目标，又作为黑盒越狱生成的对抗工具。

    Paper analysisRead analysis
  10. AttacksarXiv 2607.23496

    Concept2Scenario：用 SAE 概念归因发现可迁移的越狱场景

    提出 Concept2Scenario，借助内部概念归因发现可迁移的越狱场景

    arXiv
    2607.23496
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    White-box, Black-box
    Evaluated models
    Qwen3.5-9B, Llama-3.1-8B, Ministral-3-8B and others6

    Abstract该研究从表征空间因果归因揭示场景削弱拒绝的机制，发现可复用脆弱场景先验，提升多模型黑盒越狱效能与探索速度。

    Paper analysisRead analysis
  13. Analysis and theoryarXiv 2606.04168

    论文提出随机插入攻击，利用自回归一致性绕过 LLM 安全对齐

    解释自回归一致性如何导致浅层安全对齐，并提出随机插入攻击

    arXiv
    2606.04168
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Abstract作者用自回归一致性串起浅层对齐、插入攻击和最坏延续对齐。

    作者把浅层安全对齐和一类续写攻击都归到自回归一致性上。

    Paper analysisRead analysis
  15. AttacksarXiv 2604.24983

    PEO：直接优化原始提示词嵌入的 LLM 白盒越狱方法

    提出 PEO，直接优化原提示词嵌入实现白盒越狱

    arXiv
    2604.24983
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    White-box
    Evaluated models
    Vicuna-7B-v1.3, Llama-2-7B-Chat, Qwen3-4B-Instruct and others4
    AbstractPEO 原位优化提示词嵌入、多轮只重攻失败样本。

    PEO是一种不追加对抗 token 的 white-box 越狱：直接改已有提示词嵌入，并只对未成功样本更换续写目标再优化。

    Paper analysisRead analysis
  17. InterpretabilityarXiv 2603.08234

    续写与拒答之争：大语言模型中续写触发越狱的机制分析

    分析续写触发越狱的注意力头机制并提出 HCS

    arXiv
    2603.08234
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    LLaMA-2-7B-Chat, Qwen2.5-7B-Instruct
    Abstract作者称越狱来自两类头竞争，HCS 加蒸馏可把该竞争用于推理时干预。

    作者用注意力头机制分析 continuation-triggered jailbreak：同一有害指令下，把 continuation 后缀从指令后面挪到提示词边界外，模型更容易顺着后缀继续生成。

    Paper analysisRead analysis
Load more