Skip to content

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

Oct 12 · Monday

All papers

Found 7 papers
Filter4

Defense type

Model hardening15Inference-time intervention37Detection and filtering7Monitoring and auditing17Watermarking and provenance13

Impact

Output manipulation5

System type

Models and APIs7LLM applications21AI Agent42coding agent8web agent3Multi-agent systems5

Attack surface / entry point

Images, audio and video3

Layer

Prompt layer24Model layer7Application layer0Training and data layer13

System component

Guardrails and judges3Multimodal5
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.
Defense type：Detection and filteringSystem type：Models and APIsLayer：Model layerLayer：Application layerClear all

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

  2. 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
  3. Analysis and theoryarXiv 2609.36477

    研究发现护栏模型在基座模型不确定处过度自信

    诊断护栏相对基座在对抗提示上的置信失校

    arXiv
    2609.36477
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Llama-Guard 7B, Llama-Guard-2 8B, Llama-Guard-3 8B and others10
    Abstract对抗包装使 guard 高置信漏检，而对应 base 在同一输入上常更不确定。

    诊断性分析：五个提示词分类 guard 的置信度，在清洁输入上可以较校准，在越狱包装后则变成高置信度的错误判定。

    Paper analysisRead analysis
  4. DefensesarXiv 2608.00716

    生成图像更易被遗忘：面向合成图像检测的机器遗忘视角

    提出机器遗忘式生成图像检测，用遗忘前后特征相似度区分自然图与生成图

    arXiv
    2608.00716
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    DINOv2 ViT-S/14, DINOv2 ViT-B/14, DINOv2 ViT-L/14 and others9
    Abstract作者用剪枝或 LoRA 遗忘拉开生成图与自然图的特征相似度，并在多个检测基准上报告结果。

    这篇工作把 AI 生成图像检测写成对预训练视觉模型的机器遗忘：原模型对两类图像损失都低，剪枝后生成图特征变得更不像原模型。

    Paper analysisRead analysis
  5. DefensesarXiv 2607.14737

    GeoDetect：面向视觉语言预训练模型的几何对抗检测方法

    提出 GeoDetect 以几何分数检测视觉语言预训练模型的对抗样本

    arXiv
    2607.14737
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    White-box
    Evaluated models
    CLIPViT (ViT-B/16), CLIPCNN (ResNet-50), ALBEF and others4
    AbstractGeoDetect 用几何分数检测 VLP 对抗样本。

    GeoDetect 是面向视觉-语言预训练模型的对抗样本检测方法，用嵌入几何而不是任务 logits 区分干净样本与对抗样本。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2605.18984

    Artifact-Bench：评估 MLLM 检测与诊断 AI 生成视频伪影

    提出 Artifact-Bench，评测 MLLM 检测并诊断 AI 生成视频伪影

    arXiv
    2605.18984
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Gemini 3.1 Pro, Gemini 3 Flash, Qwen3-VL 8B-Instruct and others19
    AbstractArtifact-Bench 以三项任务评 MLLM 的 AIGC 伪影感知。

    Artifact-Bench 用来看 MLLM 能否发现 AI 生成视频中的伪影，并说明视频为何不真实。

    Paper analysisRead analysis
  7. Evaluation and benchmarksarXiv 2603.05618

    Safer Reasoning Traces: Measuring and Mitigating Chain-of-Thought Leakage in Large Language Models

    测量思维链将提示中 PII 复述进轨迹的泄漏并评测轻量检测器

    arXiv
    2603.05618
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Llama 3.3:70B, Claude Opus, Mixtral 8×22B and others7
    Abstract作者称 CoT 抬高逐字泄漏且随模型与预算而变，检测器也需按模型选择。

    在“不要复述 PII”的输出策略下，本文用 black-box 协议测量提示词中的 PII 被写回 CoT 与答案的情况，并比较轻量 gatekeeper。

    Paper analysisRead analysis
You have reached the end