Captioning on DRAMA Reasoning (val)
60.3B4 ScoreUniDrive
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| UniDriveInput=Video2026.06 | 60.3 | 39.6 | 277.5 | 58.1 | 60.8 | |
| UniDrive w/o STInput=Image2026.06 | 55.2 | 38.1 | 246.7 | 54.3 | 57.5 | |
| Video-LLAMAInput=Video2026.06 | 53.9 | 37.8 | 229.5 | 52.6 | 48.4 | |
| eP-ALMInput=Video2026.06 | 51.4 | 38 | 225.1 | 52.8 | 47.3 | |
| InstructBLIPInput=Image2026.06 | 49.9 | 37.9 | 205 | 50.9 | 48.9 | |
| Shikra*Input=Image2026.06 | 49.8 | 37.7 | 204.7 | 50.7 | 50.1 | |
| LLaVAInput=Image2026.06 | 47.5 | 35.2 | 198.6 | 48.3 | 47.4 | |
| BLIP-2Input=Image2026.06 | 46.1 | 34.3 | 194.7 | 50.7 | 46.2 |