SenBen: 用于可解释内容审核的敏感场景图
SenBen: Sensitive Scene Graphs for Explainable Content Moderation
AI总结：
提出SenBen基准和紧凑学生模型，通过多任务训练和词汇平衡策略实现敏感内容的空间定位与可解释性，在场景图生成上超越多数VLM。
摘要：
内容审核系统将图像分类为安全或不安全，但缺乏空间定位和可解释性：它们无法解释检测到了什么敏感行为、涉及谁或发生在哪里。我们引入了敏感基准（SenBen），这是第一个用于敏感内容的大规模场景图基准，包含来自157部电影的13,999帧，标注了Visual Genome风格的场景图（25个对象类别、28个属性，包括情感状态如痛苦、恐惧、攻击和痛苦，14个谓词）以及跨5个类别的16个敏感标签。我们通过多任务配方将前沿VLM蒸馏成一个紧凑的241M学生模型，该配方通过基于后缀的对象身份、词汇感知召回（VAR）损失和解耦的Query2Label标签头（带非对称损失）解决自回归场景图生成中的词汇不平衡问题，在SenBen召回率上比标准交叉熵训练提高了+6.4个百分点。在基于场景图的指标上，我们的学生模型优于除Gemini模型外的所有评估VLM和所有商业安全API，同时在所有模型中实现了最高的对象检测和字幕生成分数，推理速度提升7.6倍，GPU内存减少16倍。
英文摘要：
Content moderation systems classify images as safe or unsafe but lack spatial grounding and interpretability: they cannot explain what sensitive behavior was detected, who is involved, or where it occurs. We introduce the Sensitive Benchmark (SenBen), the first large-scale scene graph benchmark for sensitive content, comprising 13,999 frames from 157 movies annotated with Visual Genome-style scene graphs (25 object classes, 28 attributes including affective states such as pain, fear, aggression, and distress, 14 predicates) and 16 sensitivity tags across 5 categories. We distill a frontier VLM into a compact 241M student model using a multi-task recipe that addresses vocabulary imbalance in autoregressive scene graph generation through suffix-based object identity, Vocabulary-Aware Recall (VAR) Loss, and a decoupled Query2Label tag head with asymmetric loss, yielding a +6.4 percentage point improvement in SenBen Recall over standard cross-entropy training. On grounded scene graph metrics, our student model outperforms all evaluated VLMs except Gemini models and all commercial safety APIs, while achieving the highest object detection and captioning scores across all models, at $7.6\times$ faster inference and $16\times$ less GPU memory.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
