| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Spatial Reasoning | SpatialEval Spatial-Map (20% train) | Accuracy (Spatial-Map)93.4 | 20 | |
| Spatial Reasoning | SpatialEval Spatial-Grid (20% train) | Accuracy92.2 | 20 | |
| Spatial Reasoning | SpatialEval Maze-Nav 20% (train) | Accuracy86.2 | 20 | |
| Spatial Reasoning | SpatialEval (test) | Maze Navigation Acc35.2 | 16 | |
| Topological Classification | SpatialEval US Boundaries | PG-PG Topological Accuracy97.3 | 12 | |
| Distance Estimation | SpatialEval US Boundaries | PG-PG Distance0.37 | 12 | |
| Nearest Neighbor Query | SpatialEval Los Angeles | PT-PL NN Accuracy95.1 | 12 | |
| Distance Estimation | SpatialEval Los Angeles | PT-PL Distance543.6 | 12 | |
| Nearest Neighbor Query | SpatialEval Beijing | PT-PL NN Accuracy95.7 | 12 | |
| Distance Estimation | SpatialEval Beijing | PT-PL Distance393.5 | 12 | |
| Spatial Reasoning | SpatialEval | Accuracy70.81 | 12 | |
| Polygon Area Estimation | SpatialEval US Boundaries 1.0 (test) | MAE0.055 | 10 | |
| Polyline Length Estimation | SpatialEval Los Angeles 1.0 (test) | MAE38.9 | 10 | |
| Polygon Area Estimation | SpatialEval Los Angeles 1.0 (test) | MAE884.4 | 10 | |
| Polyline Length Estimation | SpatialEval Beijing 1.0 (test) | MAE60.6 | 10 | |
| Polygon Area Estimation | SpatialEval Beijing 1.0 (test) | MAE1,050.9 | 10 | |
| Spatial Reasoning (Single-Image) | SpatialEval Real | Accuracy68.9 | 10 |