| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| SPAR-Bench-MV + MindCube-Tiny + MMSI-Bench (test) | GPT-5.2 | Overall Score51 | 32 | 23d ago | |
| SPBench-MV | GAMSI_S1+S2 | Accuracy80.4 | 19 | 2mo ago | |
| MindCube-Tiny (test) | GPT-5.2 | Aro Accuracy69.2 | 17 | 5mo ago | |
| MMSI | Gemini-2.5-Pro | Overall Accuracy38 | 15 | 5mo ago | |
| SPAR-Bench-MV (test) | GPT-5.2 | Score (Low Difficulty)43.7 | 15 | 5mo ago |