Skip to content

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

Oct 12 · Monday

All papers

Found 11 papers
Filter4

Model risks

Inappropriate refusal3Hallucination and fabrication6

Impact

Output manipulation8

System type

Models and APIs10

Layer

Prompt layer14Model layer11Application layer12Training and data layer3

System component

Guardrails and judges5Monitors5Reasoning chains6Vision8Audio4Video5Image generation3Embeddings0Multimodal16
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 232papers have no tags yet and are not included in filtered results.
Layer：Model layerSystem component：VisionSystem component：Image generationSystem component：EmbeddingsClear all

Another 232papers do not have category tags yet and are not included in filtered results.

  4. Evaluation and benchmarksarXiv 2609.32763

    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

    arXiv
    2609.32763
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    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
  5. Evaluation and benchmarksarXiv 2609.06991

    AV-SafetyBench：面向文本到音视频生成的安全基准

    提出 AV-SafetyBench 评测文生音视频模型的不安全生成

    arXiv
    2609.06991
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    LTX-2.3, Ovi-1.1, JavisDiT++ and others7
    AbstractAV-SafetyBench 用三视角评五个 T2AV 模型，显示只看视频会漏掉音频与联合风险。

    AV-SafetyBench 是面向文本生成音视频（T2AV）的安全基准：一条提示词同时规定画面和音频，评测同时看两轨以及它们的交互。作者要测的是，不安全含义不只出现在画面上：它可以主要由语音或声音承载，也可以只有两轨合起来才成立。T2VSafetyBench 不评音频，TTA-Bench 不评与画面的交互，作者认为二者都盖不住这些路径。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2609.04720

    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

    arXiv
    2609.04720
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    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
  7. Evaluation and benchmarksarXiv 2609.22206

    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

    arXiv
    2609.22206
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    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
  8. Evaluation and benchmarksarXiv 2608.11816

    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

    arXiv
    2608.11816
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    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
  9. Evaluation and benchmarksarXiv 2606.24115

    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

    arXiv
    2606.24115
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    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
  10. Evaluation and benchmarksarXiv 2604.12447

    HazardArena：面向 VLA 模型的语义安全评测基准

    提出 HazardArena 基准，用孪生场景评测 VLA 的语义安全

    arXiv
    2604.12447
    Published
    Layer
    Model layer
    Setting
    Embodied systems and robotics
    Evaluated models
    OpenVLA-OFT, π0, NORA and others4
    Abstractsafe-only 微调提升良性执行，也提升匹配不安全孪生上的危险完成。

    HazardArena 评测 VLA 在运动要求不变时，能否因语义风险而停手。

    Paper analysisRead analysis
You have reached the end