Image Classification on ImageNet Variants (V2, R, A, O, Sketch) (test)
37.5IN-V2 Top-1 AccCLIMP
Evaluation Results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CLIMPVision Tower=VMamba-B, Text Tower=Mamba-12026.01 | 37.5 | 68.4 | 46.2 | 74.5 | 15.5 | 45.3 | 48.1 | 78.8 | 27.5 | 54.4 | 35.2 | 64.3 | |
| CLIMPVision Tower=VMamba-B, Text Tower=Mamba-22026.01 | 37 | 67.6 | 46.6 | 74.5 | 15.6 | 45.5 | 49.8 | 77 | 27 | 54.2 | 34.8 | 63.7 | |
| RoPE-ViT-B/16Vision Tower=RoPE-ViT-B/16, Text Tower=LLaMA-3.22026.01 | 34.4 | 65.4 | 47.8 | 76 | 16.3 | 46.3 | 40.1 | 69.9 | 27.4 | 53.1 | 33.2 | 62.1 | |
| FlexViT-B/16Vision Tower=FlexViT-B/16, Text Tower=LLaMA-3.22026.01 | 32.8 | 62.5 | 45.4 | 74.3 | 13.2 | 41.2 | 38.1 | 68.2 | 24.6 | 50 | 30.8 | 59.2 | |
| NaFlex-B/16Vision Tower=NaFlex-B/16, Text Tower=LLaMA-3.22026.01 | 31.4 | 60.5 | 44.8 | 72.5 | 12 | 36 | 38.5 | 67.6 | 23.5 | 49.1 | 30 | 57.1 | |
| ViT-B/16Vision Tower=ViT-B/16, Text Tower=LLaMA-3.22026.01 | 30 | 57.5 | 42.9 | 69.7 | 13.9 | 37.6 | 27.7 | 55.8 | 21.7 | 46.1 | 27.2 | 53.3 |