Video Classification on Kinetics (test val)
78.1Top-1 AccuracyX-Ray Visual EViT-2b
Evaluation results
| Method | Links | ||
|---|---|---|---|
| X-Ray Visual EViT-2bEncoder=EViT-2b, Evaluation Protocol=Linear probing2026.02 | 78.1 | 98.37 | |
| PE-GEncoder=PE-G, Evaluation Protocol=Linear probing2026.02 | 76.68 | 92.79 | |
| PE-LEncoder=PE-L, Evaluation Protocol=Linear probing2026.02 | 72.1 | 90.32 | |
| DiNO-v3Encoder=7B, Evaluation Protocol=Linear probing2026.02 | 68.1 | 84.51 |