Image Classification on UTO → LAB
88.35Average AccuracyInfoNCE (I+P→I+P)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| InfoNCE (I+P→I+P)Model=EfficientNet B02026.03 | 88.35 | 0.7 | |
| InfoNCE (I+P→I+P)Model=ViT Tiny2026.03 | 88.3 | 0.71 | |
| InfoNCE (I+P→I+P)Model=ConvNeXt Femto2026.03 | 88.13 | 0.74 | |
| InfoNCE (I→I)Model=EfficientNet B02026.03 | 69.87 | 1.21 | |
| InfoNCE (I→I)Model=ViT Tiny2026.03 | 66.11 | 1.22 | |
| InfoNCE (I→I)Model=ConvNeXt Femto2026.03 | 66.03 | 1.13 | |
| DINOModel=EfficientNet B02026.03 | 57.24 | 0.81 | |
| DINOModel=ConvNeXt Femto2026.03 | 54.76 | 0.87 | |
| DINOModel=ViT Tiny2026.03 | 53.24 | 1.01 |