Skip to content

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

Oct 12 · Monday

All papers

Found 16 papers
Filter4

Attack type

Jailbreaks9

Attack technique

Gradient and representation optimization3Cross-modal payloads5Defense-targeted3

Impact

Harmful content9Output manipulation4

System type

Models and APIs14

Attacker access

Black-box5

Attack surface / entry point

User input7Images, audio and video10

Layer

Prompt layer15Model layer36Application layer66Training and data layer12Infrastructure layer1

System component

Guardrails and judges63Monitors5Reasoning chains15Fine-tuning and open weights3Vision60Audio11Video6Image generation18Multimodal91
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 233papers have no tags yet and are not included in filtered results.
Layer：Prompt layerLayer：Infrastructure layerSystem component：MonitorsSystem component：AudioClear all

Another 233papers do not have category tags yet and are not included in filtered results.

  2. Analysis and theoryarXiv 2610.07005

    研究审计 Qwen2-Audio 音频越狱 AdvWave-P 的频率与解码层归因

    遮挡审计 AdvWave-P 音频越狱的频带必要性与解码层关联

    arXiv
    2610.07005
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Qwen2-Audio-7B-Instruct, LLaMA-Omni
    AbstractAdvWave-P 的八频带排序与能量相关。

    这是对加性音频越狱 AdvWave-P 的频率与解码器深度审计，受害模型主要是 Qwen2-Audio-7B-Instruct，不提出新攻击。

    Paper analysisRead analysis
  3. AttacksarXiv 2609.15989

    Researchers propose plan injection attack that can bypass chain-of-thought monitoring

    提出 plan injection 攻击，植入良性伪装计划以绕过思维链监视器

    arXiv
    2609.15989
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Gray-box
    Evaluated models
    Qwen2.5-Coder-7B-Instruct, Qwen3-8B, DeepSeek-R1-0528 and others6

    Abstract论文指出模型会将外部上下文中的伪装计划改写为自身推理，使思维链监视器失效甚至反向合理化恶意行为。

    Paper analysisRead analysis
  4. DefensesarXiv 2609.05797

    Paper finds safety monitors mainly block requests the model would already refuse

    评估安全监视器并提出双侧梯级微调以减少会答有害请求的漏检

    arXiv
    2609.05797
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Gemma-4-31B-IT, Qwen3-32B, Llama Guard 3 8B and others6

    Abstract论文发现安全监视器大多拦截模型已拒答请求，指出明确度偏差是共性原因，并通过双侧梯级微调提升回答请求的召回表现。

    Paper analysisRead analysis
  5. AttacksarXiv 2609.01023

    Akrasia：针对推理型代码 LLM 的隐蔽后门攻击

    提出隐蔽后门攻击 AKRASIA，在推理代码模型提示中植入触发器并伪装思维链

    arXiv
    2609.01023
    Published
    Layer
    Prompt layer
    Setting
    coding agent
    Evaluated models
    Claude-Sonnet-5, GPT-5.5, Gemini-3.5-flash and others7

    Abstract论文针对推理代码大模型提出了隐蔽推断期后门攻击 AKRASIA，揭示了模型思维链可被利用进行欺骗性伪装的安全隐患。

    Paper analysisRead analysis
  7. Evaluation and benchmarksarXiv 2606.06037

    SpeechJBB：语码转换语音下大型音频语言模型的越狱基准

    提出 SpeechJBB，评测语码转换语音下音频语言模型的越狱稳健性

    arXiv
    2606.06037
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Qwen2.5-Omni-7B, Qwen3-Omni-30B-A3B-Instruct, Voxtral-Small-24B and others9
    AbstractSpeechJBB 用五语语码转换语音和伪词探针评测九个 LALM，作者将越狱归为对齐失败而非听不懂。

    SpeechJBB 用合成语音测量 LALM 在五语、语码转换和局部伪词混淆下的安全对齐，并另设口语理解对照。

    Paper analysisRead analysis
  8. AttacksarXiv 2605.30040

    研究揭示 LLM 按 token 计费可被服务商系统性虚报

    构造可通过计费审计的 token 虚报，使服务商系统性超收

    arXiv
    2605.30040
    Published
    Layer
    Infrastructure layer
    Setting
    Models and APIs
    Evaluated models
    PALACE domain-adapted auditor
    Abstract不诚实提供方可在三种 token 审计上虚报并通过检查，作者认为需改用提供方不控制的证据。

    不透明 LLM 的按 token 计费里，现有审计核验的是提供方可控的证据，不诚实提供方可以虚报并通过检查。。

    Paper analysisRead analysis
  11. AttacksarXiv 2604.14604

    The AudioHijack framework achieves covert audio prompt injection against 13 speech large models

    提出 AudioHijack，用不可感知对抗音频对语音模型做上下文无关提示注入

    arXiv
    2604.14604
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    SpeechGPT, GLM-4-Voice, VITA-Audio and others15

    Abstract提出了 AudioHijack 框架，揭示了 LALM 对听觉提示注入的脆弱性与传统防御的局限。

    Paper analysisRead analysis
  14. AttacksarXiv 2511.10692

    StyleBreak：利用风格化音频越狱揭示大型音频语言模型的对齐漏洞

    提出 StyleBreak 风格化音频越狱，利用语音属性绕过音频语言模型对齐

    arXiv
    2511.10692
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Qwen2-Audio-7B-Instruct, MERaLiON-AudioLLM-Whisper-SEA-LION, Ultravox-v0.4.1-Llama-3.1-8B and others6
    Paper analysisRead analysis
  15. AttacksarXiv 2504.01094

    研究者提出 Multi-AudioJail

    提出 MULTI-AUDIOJAIL，以多语多口音和声学扰动越狱音频模型

    arXiv
    2504.01094
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Qwen2-Audio, DiVA-llama-3-v0-8b, MERaLiON-AudioLLM-Whisper-SEA-LION and others5

    Abstract黑盒多语多口音音频加声学扰动可抬高 LALM 的 JSR，文本示范防御并非在每个模型上都下降。

    Paper analysisRead analysis
You have reached the end