Scientific Reasoning on MMK12-Bio
65.7AccuracyV-STAR
Evaluation Results
| Method | Links | |
|---|---|---|
| V-STARData Size=40k2026.04 | 65.7 | |
| OpenVLThinkerData Size=59.2k2026.04 | 65 | |
| VL-RethinkerData Size=39k2026.04 | 64.8 | |
| ThinkLite-VLData Size=11k2026.04 | 62.9 | |
| VL-CogitoData Size=80k2026.04 | 61.3 | |
| Vision-R1Data Size=210k2026.04 | 57.9 | |
| Qwen2.5VL2026.04 | 54 | |
| R1-OnevisionData Size=155k2026.04 | 40.8 |