Skip to content

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

Oct 12 · Monday

All papers

Found 5 papers
Filter5

Attack type

Jailbreaks32Prompt injection6Poisoning and backdoors8Extraction and theft2Detection evasion8Malicious use3

Impact

Data and privacy leaks3

System type

Models and APIs5coding agent0

Attack surface / entry point

Images, audio and video4

System component

Reasoning chains5Fine-tuning and open weights5Vision5Audio3Image generation11Multimodal19
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 type：Extraction and theftAttack type：Malicious useSystem type：Models and APIsSystem type：coding agentSystem component：VisionClear all

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

  1. Evaluation and benchmarksarXiv 2609.33481

    IDUnlearn-Bench：面向视觉语言模型的个体级遗忘基准

    提出 IDUnlearn-Bench，评测 VLM 个体级遗忘后的身份残留

    arXiv
    2609.33481
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-VL-2B, Qwen3-VL-4B, Qwen3-VL-8B and others6
    AbstractIDUnlearn-Bench 用四族探针显示：压低直接属性访问后，身份仍常能被反向查出、绑定或重构。

    IDUnlearn-Bench 评测 VLM 是否忘掉一个可跨模态识别的人，而不只是答不出某条监督事实。

    Paper analysisRead analysis
  3. DefensesarXiv 2609.15671

    QPriv-VL：面向视觉语言模型的问题引导 token 裁剪隐私防御

    提出 QPriv-VL，用问题引导剪枝降低分布式视觉问答的隐私泄露

    arXiv
    2609.15671
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    LLaVA-1.5-7B, Llama-2
    AbstractQPriv-VL 在切层前按问题相关度与 DINOv2 敏感度剪视觉 token，作者称能在约四成 token 下兼顾准确率与四类攻击。

    QPriv-VL 把问题引导的视觉 token 剪枝做成分布式 VQA 的客户端隐私防御，而不是只做集中式推理加速。

    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. AttacksarXiv 2601.12983

    ChartAttack：测试大语言模型在图表生成中对恶意提示的脆弱性

    提出 ChartAttack，用恶意提示驱动代码模型生成误导图表

    arXiv
    2601.12983
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Claude-4.6-Sonnet, GPT-4o, Gemini-2.5-Flash and others20
    AbstractChartAttack 生成不改数据的误导图，降低 MLLM 与人类的图表问答准确率。

    ChartAttack 研究恶意提示能否让 MLLM 在不改数据的情况下生成误导图表，并量化这些图对其他 MLLM 和人类图表问答的影响。

    Paper analysisRead analysis
You have reached the end