测试模型对嵌套闭合曲线拓扑关系的精确推理能力，揭示当前视觉模型仍难胜任复杂空间逻辑。
CurveBench: A Benchmark for Exact Topological Reasoning over Nested Jordan Curves
- 基于756张非相交闭合曲线图像构建层级拓扑推理任务
- 最强模型在难题上准确率仅19.1%，显示视觉推理仍有巨大差距
- 适合研究视觉-语言模型拓扑感知能力的学者使用
我们提出CurveBench，一个基于视觉输入的层次化拓扑推理基准。该基准包含756张图像，涵盖简单、多边形、地形启发、迷宫式和密集计数五种配置下的非相交乔丹曲线。每张图像均标注了编码平面区域包含关系的有根树。任务设定为结构化预测：给定图像，模型需恢复由曲线诱导的完整有根包含树。尽管任务视觉上简单，最强评估模型Gemini 3.1 Pro在CurveBench-Easy上仅达71.1%准确率，在CurveBench-Hard上仅为19.1%。我们通过类似RLVR的微调验证基准实用性，训练后的Qwen3-VL-8B模型在CurveBench-Easy上从2.8%提升至33.3%，超过GPT-5.4与Claude Opus 4.5。剩余差距，尤其在CurveBench-Hard上，表明精确拓扑感知的视觉推理仍未解决。
原文摘要 · Abstract (English)
We introduce CurveBench, a benchmark for hierarchical topological reasoning from visual input. CurveBench consists of \textbf{756 images} of pairwise non-intersecting Jordan curves across easy, polygonal, topographic-inspired, maze-like, and dense counting configurations. Each image is annotated with a rooted tree encoding the containment relations between planar regions. We formulate the task as structured prediction: given an image, a model must recover the full rooted containment tree induced by the curves. Despite the visual simplicity of the task, the strongest evaluated model, Gemini 3.1 Pro, achieves only \textbf{71.1\%} tree-generation accuracy on CurveBench-Easy and \textbf{19.1\%} on CurveBench-Hard. We further demonstrate benchmark utility through RLVR-style fine-tuning of open-weight vision-language models. Our trained Qwen3-VL-8B model improves over \texttt{Qwen-3-VL-8B-Thinking} from \textbf{2.8\%} to \textbf{33.3\%} tree-generation accuracy on CurveBench-Easy, exceeding GPT-5.4 and Claude Opus 4.5 under our evaluation protocol. The remaining gap, especially on CurveBench-Hard, shows that exact topology-aware visual reasoning remains far from solved.
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。