| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| VSR | Qwen3-VL-32B | Accuracy88.87 | 59 | 14d ago | |
| SpaceVista-Bench Overall | SpaceVista-7b | Score39.2 | 20 | 7d ago | |
| SpaceVista-Bench Outdoor | Score43 | 20 | 7d ago | ||
| SpaceVista-Bench Indoor | VLM-3R | Score45.1 | 20 | 7d ago | |
| SpaceVista-Bench Tabletop | Score38.7 | 20 | 7d ago | ||
| What's Up (Split B) | Accuracy100 | 20 | 3mo ago | ||
| What's Up (Split A) | Accuracy100 | 20 | 3mo ago | ||
| Visual Spatial Reasoning | InternVL3.5-8B | Accuracy79.3 | 16 | 15d ago | |
| SceneGraphs (test) | NAG-LoRA | Accuracy83.82 | 7 | 3mo ago | |
| VSR ZOOM-Hard | Mags-RL | GPT Accuracy55.98 | 6 | 6d ago | |
| VSR ZOOM-Medium | Mags-RL | GPT Accuracy67.63 | 6 | 6d ago | |
| VSR (ZOOM-Easy) | Mags-RL | GPT Accuracy73.09 | 6 | 6d ago | |
| GWR set (test) | SpaceVista-7B | Accuracy31.1 | 4 | 7d ago | |
| SpaceVista-Bench size-related | SpaceVista-7B | Accuracy58.3 | 4 | 7d ago | |
| MMSI-Bench | Cam-Cam Accuracy36.6 | 4 | 3mo ago |