| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Geometry3K (val) | PEPO_D | Accuracy28.12 | 12 | 24d ago | |
| Q-Spatial++ | TIGeR | δ≤2 Score70.3 | 9 | 1mo ago | |
| Geometry3k (test) | DS (Oracle) | Test Score48.11 | 8 | 1mo ago | |
| Geometry3K | GPT-4o + Sketchpad | Geometry Score66.7 | 7 | 1mo ago | |
| GeoEval (val) | CalibRL | Accuracy33.44 | 6 | 1mo ago |