Skip to content

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

Oct 13 · Tuesday

All papers

Found 13 papers
Filter3

Attack technique

Gradient and representation optimization3

Model risks

Hallucination and fabrication3

Impact

Harmful content11Data and privacy leaks4Output manipulation10Availability and reliability3

System type

Models and APIs11

Attack surface / entry point

Images, audio and video8

Layer

Prompt layer3Model layer6

System component

MCP and skills11Memory26RAG31Guardrails and judges43Monitors7Reasoning chains18Permissions and sandboxing9Fine-tuning and open weights18Vision58Audio13Video9Image generation10Embeddings6Multimodal85
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 215papers have no tags yet and are not included in filtered results.
Impact：Output manipulationImpact：Availability and reliabilitySystem component：AudioClear all

Another 215papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2610.09819

    FAB 攻击：给 HuBERT/WavLM 植入后门，背景警笛声即可让下游语音任务失灵

    提出 FAB 攻击，向声学基础模型植入可由环境声触发的后门

    arXiv
    2610.09819
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    HuBERT-base, WavLM-base

    Abstract论文提出了针对声学基模的 FAB 后门攻击，在弱威胁模型下实现了任务无关、物理可触发的后门植入。

    Paper analysisRead analysis
  3. Evaluation and benchmarksarXiv 2610.06339

    BabelFake：面向多语言音视频 DeepFake 检测的基准数据集

    构建多语言音视频 DeepFake 基准 BabelFake 并评测检测器

    arXiv
    2610.06339
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    SpeechForensics, AVH-Align, (B)RAVEn
    AbstractBabelFake 为五语同意采集基准。

    BabelFake 是经同意与 IRB 批准采集的多语音视频 DeepFake 检测基准，用来把视觉操纵和语音合成拆开比较。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2609.38106

    剪枝提效却牺牲公平：剪枝语音 LLM 中的群体差异

    测量剪枝对语音 LLM 群体转写公平性的影响

    arXiv
    2609.38106
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Whisper Small (12 layers), Whisper Medium (24 layers), Whisper Large-v2 (32 layers)

    Abstract剪枝后的 Speech-LLM 不能只看总体 WER：Fair-Speech 上族群差扩大，作者建议看最差组。

    Paper analysisRead analysis
  7. DefensesarXiv 2609.37568

    SECRET：用源条件化中继引导缓解音视频大语言模型的幻觉

    提出训练无关的 SECRET，对问题表征做源条件转向以缓解音视频模型幻觉

    arXiv
    2609.37568
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    VideoLLaMA2-AV-7B, Qwen2.5-Omni-7B, Qwen3-Omni-30B-A3B
    AbstractSECRET 在问题状态上做源条件转向，三个 AVLLM 上减轻了跨模态接地幻觉。

    SECRET 是一种训练无关的问题状态转向方法，用来减轻 AVLLM 的 source-confused grounding hallucination。

    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
  12. AttacksarXiv 2602.07379

    Aegis: Governance, Integrity, and Security for AI Voice Agents

    Proposes the Aegis red-teaming framework for multi-turn attacks on high-risk voice agents.

    arXiv
    2602.07379
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-4o, GPT-4o-mini, Gemini-1.5 Pro and others7
    AbstractAegis red-teams three types of voice agents: query-based access eliminates authentication and leakage successes, but behavioral attacks remain.

    Aegis uses multi-turn red-teaming evaluation on ALLM voice agents already integrated into business systems. The authors call this the first systematic red-teaming framework in this setting.

    Paper analysisRead analysis
You have reached the end