让AI学会看懂图像中事物的因果关系，超越单纯识别。
CauSight: Learning to Supersense for Visual Causal Discovery
- 基于3.2万张带因果图标注的图像，训练模型理解视觉实体间的因果链。
- 在因果推理任务上超越GPT-4.1，准确率提升21个百分点。
- 适合研究因果推理、多模态理解或可解释AI的开发者与学者。
因果思维使人类不仅能感知所见，还能理解其发生原因。为在现代AI系统中复现此能力，我们提出视觉因果发现任务——要求模型在多样化场景中推断视觉实体间的因果关系，而非仅感知其存在。为此，我们构建了包含超过32,000张图像的视觉因果图数据集（VCG-32K），并开发了CauSight模型，通过因果感知推理实现视觉因果发现。训练方法融合三要素：(1) 基于VCG-32K的数据筛选；(2) 树状因果思维（ToCT）生成推理路径；(3) 设计因果奖励的强化学习优化推理策略。实验表明，CauSight在视觉因果发现任务中优于GPT-4.1，性能提升超过三倍（绝对增益21%）。代码、模型与数据集已开源，详见项目主页：https://github.com/OpenCausaLab/CauSight。
原文摘要 · Abstract (English)
Causal thinking enables humans to understand not just what is seen, but why it happens. To replicate this capability in modern AI systems, we introduce the task of visual causal discovery. It requires models to infer cause-and-effect relations among visual entities across diverse scenarios instead of merely perceiving their presence. To this end, we first construct the Visual Causal Graph dataset (VCG-32K), a large-scale collection of over 32,000 images annotated with entity-level causal graphs, and further develop CauSight, a novel vision-language model to perform visual causal discovery through causally aware reasoning. Our training recipe integrates three components: (1) training data curation from VCG-32K, (2) Tree-of-Causal-Thought (ToCT) for synthesizing reasoning trajectories, and (3) reinforcement learning with a designed causal reward to refine the reasoning policy. Experiments show that CauSight outperforms GPT-4.1 on visual causal discovery, achieving over a threefold performance boost (21% absolute gain). Our code, model, and dataset are fully open-sourced at project page: https://github.com/OpenCausaLab/CauSight.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。