CurveBench: A Benchmark for Exact Topological Reasoning over Nested Jordan Curves
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\...
| Authors | Amirreza Mohseni, Mona Mohammadi, Morteza Saghafian et al. |
|---|---|
| Primary category | cs.CV |
| Published | 13 May 2026 |
| Last updated at source | 15 May 2026 |
Models and datasets referencing this paper
- AmirMohseni/CurveBench-EasyDataset
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.