Scientific Reasoning on MMK12-Phys
56.9AccuracyV-STAR
Evaluation Results
| Method | Links | |
|---|---|---|
| V-STARData Size=40k2026.04 | 56.9 | |
| OpenVLThinkerData Size=59.2k2026.04 | 53.8 | |
| Vision-R1Data Size=210k2026.04 | 47.3 | |
| VL-RethinkerData Size=39k2026.04 | 47.2 | |
| ThinkLite-VLData Size=11k2026.04 | 46.9 | |
| Qwen2.5VL2026.04 | 45.4 | |
| VL-CogitoData Size=80k2026.04 | 43.2 | |
| R1-OnevisionData Size=155k2026.04 | 33.8 |