Skip to content

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

Oct 12 · Monday

All papers

Found 19 papers

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

  1. 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

    Published
    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
  2. DIBench:面向移动 GUI 智能体多候选选择决策完整性的安全基准

    用 DIBench 评测移动 GUI 智能体被 UI 注入后的决策完整性

    Published
    Evaluated models
    Qwen2.5-VL(32B), Qwen3-VL(32B), GUI-Owl(32B) and others7

    Abstract论文提出移动智能体决策完整性基准 DIBench,发现欺诈 UI 注入诱导目标选择并推高完成率,通用防御收益不稳定。

    Paper analysisRead analysis
  3. Mandela-Bench: 36 multimodal models rely more on memory than on observing the image

    构建 Mandela-Bench,评测多模态模型是否凭记忆接受篡改图像

    Published
    Evaluated models
    Gemini-3.8-Flash, GPT-5.6, GLM-5.3-Flash and others15
    Abstract论文构建 Mandela-Bench 发现模型对经典图像存在记忆重构偏见,过度依赖先验导致无法有效利用知识核验事实篡改。

    该研究系统评估了多模态大语言模型在面对无痕事实性编辑时,能否利用自身内在世界知识核验被识别出的经典图像。

    Paper analysisRead analysis
  4. Knowing What Not to Answer: Selective Non-Compliance in Vision-Language Models

    Propose the KoNA benchmark to evaluate selective non-compliance in vision-language models

    Published
    Evaluated models
    InternVL3-2B-Instruct, InternVL3-78B-Instruct, Qwen2.5-VL-3B-Instruct and others8
    AbstractKoNA evaluates selective noncompliance with five types of paired queries.

    KoNA is used to evaluate whether vision-language models can selectively refuse mixed queries: only handle the components that should not be followed, while answering the remaining components.

    Paper analysisRead analysis
  5. 多模态大语言模型免训练不确定性估计的系统研究

    比较多模态大模型上三类免训练不确定性估计

    Published
    Evaluated models
    InternVL2-2B, InternVL2-8B, InternVL3.5-2B and others8
    Abstract回答长度决定免训练 UE 族。

    这是一项对 MLLM 免训练不确定性估计的分组比较:幻觉检测被作者视为弃答、转交或路由的一种用途,而不是唯一目标。

    Paper analysisRead analysis
  6. Study audits 9 vision-language models: national stance rewriting rate under Chinese prompts is about three times that under English

    审计视觉语言模型在敏感图像上从拒答转向国家立场重构的行为

    Published
    Evaluated models
    Qwen2-VL-7B, Qwen2.5-VL-7B, Qwen3-VL-8B and others9
    Abstract论文系统审计了 VLM 多模态审查,发现审查正从显式拒答转向隐蔽重构,且中文提示与文本先验会放大重构倾向。

    论文系统研究了中外开源视觉语言模型在政治敏感图像上的审查行为,主要内容如下。

    Paper analysisRead analysis
  7. 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

    Published
    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
  8. DefensesarXiv 2607.24017

    多模态大模型注意力中的语义与结构偏差解耦:SPAR 免训练干预方法

    提出 SPAR 免训练注意力干预,削弱多模态模型的结构偏置

    Published
    Evaluated models
    LLaVA-1.5-7B, LLaVA-NeXT
    AbstractSPAR 以结构偏置净化缓解视觉幻觉,结果需区分误导后收益与原始任务例外。

    SPAR 是一种无需训练、直接修改多模态内部注意力的幻觉防御。 作者关注视觉信息被查询不变的注意力偏置稀释,以及误导文本进一步推动模型依赖语言先验的问题。

    Paper analysisRead analysis
  9. 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

    Published
    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
  10. Gut-VLM 基准:九种幻觉检测方法在胃肠道内镜 VLM 上的评测

    在 Gut-VLM 上评测内镜 VLM 的幻觉检测方法

    Published
    Evaluated models
    MedGemma-4B, MedGemma-27B, LLaVA-Med-7B and others5
    Abstract九种检测方法在 Gut-VLM 五模型上的比较中,ReXTrust 的 AUC 均为最高。

    这是一个胃肠道内镜 VQA 幻觉检测的比较基准,不是新检测器结构的提出。

    Paper analysisRead analysis
  11. AttacksarXiv 2606.12716

    PaperGuard:针对多模态 AI 同行评审的攻击与防御基准

    提出 PaperGuard,用图文注入与梯度扰动抬高自动审稿分数并检测恶意指令

    Published
    Evaluated models
    Claude-sonnet-4.5, GPT-4o, GPT-4.1 and others11
    AbstractPaperGuard 测到提示注入与白盒视觉抬分,并比较分块检索防御。

    PaperGuard同时评测并检测多模态 AI 审稿中的评分操纵。

    Paper analysisRead analysis
  12. AttacksarXiv 2603.16960

    Adversarial attacks against modern vision-language models

    Conduct adversarial attacks on product images for open-source visual shopping agents and induce wrong purchases

    Published
    Setting
    web agent
    Evaluated models
    LLaVA-v1.5-7B, Qwen2.5-VL-7B
    AbstractThe authors state that LLaVA is easily induced by three attacks to buy the wrong item, while Qwen's correct purchase rate remains close to its clean condition.

    The authors compare the robustness of two open-source VLM shopping agents against image adversarial perturbations in a self-built e-commerce environment.

    Paper analysisRead analysis
You have reached the end