Image Classification on ImageNet-R 10 tasks
90.5ACC10TRM
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TRMBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 90.5 | — | |
| MagMAXBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 89.4 | — | |
| BECAMEBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 87.8 | — | |
| TIESBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 87.4 | — | |
| Model StockBackbone=ViT-L/14, Pre-training=WebImageText2026.05 | 86.8 | — | |
| TRMBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 83.1 | — | |
| MagMAXBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 80.4 | — | |
| BECAMEBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 80 | — | |
| TIESBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 79.3 | — | |
| Model StockBackbone=ViT-B/16, Pre-trained dataset=LAION-400M2026.05 | 77 | — | |
| InfLoRAPre-trained=iBOT-1k2024.03 | 71.84 | 78.29 | |
| HiDe-PromptPre-trained=iBOT-1k2024.03 | 71.33 | 73.62 | |
| InfLoRAPre-trained=DINO-1k2024.03 | 68.31 | 76.15 | |
| HiDe-PromptPre-trained=DINO-1k2024.03 | 68.11 | 71.7 | |
| SeqLORAPre-trained=iBOT-1k2024.03 | 66.87 | 71.8 | |
| SeqLORAPre-trained=DINO-1k2024.03 | 60.67 | 66.29 |