Image Classification on ImageNet-R 5 tasks
90.8AccuracyTRM
Evaluation results
| Method | Links | |
|---|---|---|
| TRMBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 90.8 | |
| MagMAXBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 89.7 | |
| BECAMEBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 87.9 | |
| TIESBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 87.8 | |
| Model StockBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 87.2 | |
| TRMBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 83.4 | |
| MagMAXBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 82.2 | |
| TIESBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 81.7 | |
| BECAMEBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 81.2 | |
| Model StockBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 79.6 |