Skip to content

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

Oct 12 · Monday

All papers

Found 9 papers
Filter6

Defense type

Model hardening3

System type

Models and APIs6LLM applications3AI Agent3coding agent0

Attack surface / entry point

Images, audio and video3

Layer

Prompt layer11Model layer15Application layer3Training and data layer6

System component

MCP and skills82Memory77RAG9Guardrails and judges66Monitors64Reasoning chains27Permissions and sandboxing90Fine-tuning and open weights53Vision39Audio9Image generation9Multimodal57
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 230papers have no tags yet and are not included in filtered results.
System type：Models and APIsSystem type：AI AgentSystem type：coding agentLayer：Application layerLayer：Training and data layerSystem component：AudioClear all

Another 230papers 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 2609.35922

    VoxParity 评测：语音智能体在需要听声判断时仍按文字行事

    提出 VoxParity，固定转写只改音频，评测语音智能体是否按行业规则改工具调用

    arXiv
    2609.35922
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    MiMo-V2.6-Pro, gemini-3.7-flash, Qwen3.8-Omni and others15
    AbstractVoxParity 用固定转写的音频对比和 words-only null，测量语音智能体是否按所听内容改变工具调用。

    VoxParity 测量语音智能体在词语不变、只有声音改变时，是否按行业规则改它所执行的工具调用。

    Paper analysisRead analysis
  5. DefensesarXiv 2609.12623

    SteerDuplex：可操控的全双工语音对话模型

    提出 SteerDuplex，用监督微调与两阶段强化学习提升全双工语音对话可控性

    arXiv
    2609.12623
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    SteerDuplex-SFT, SteerDuplex-RL
    AbstractSteerDuplex 用 SFT 提升语音可控，用两阶段 RL 改时序，并记录奖励黑客。

    SteerDuplex 是基于 Moshi 的全双工语音模型，目标是在保留轮替、打断和一般任务能力的同时，按用户指令改变内容与发声。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2609.09404

    MMPIBench：多模态提示注入对六种 Agent 框架的跨框架评测

    用 MMPIBench 评测多模态提示注入对 Agent 框架的影响

    arXiv
    2609.09404
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Claude Opus 4.8, GPT-5.4, Gemini 3.1 Pro and others6

    Abstract论文提出 MMPIBench 评测多模态提示注入对智能体的影响，发现模型主导行为、尝试远超完成，音频通道缺乏防护。

    Paper analysisRead analysis
  8. DefensesarXiv 2608.18607

    VA-Judger：面向联合视频-音频生成的人类偏好反馈奖励建模

    提出 VA-Judger，用人类偏好训练联合音视频奖励模型

    arXiv
    2608.18607
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    VA-Judger, LTX-2 + VA-Judger
    AbstractVA-Judger 用易到难的人类偏好学习做联合音视频奖励，并用于 LTX-2 后训练。

    VA-Judger 是面向联合音视频生成的思维链全模态奖励模型，用来替代由分维度专家指标拼成的强化学习奖励。

    Paper analysisRead analysis
  9. AttacksarXiv 2602.07379

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

    提出 Aegis 红队框架，多轮攻击高风险语音 Agent

    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