Image Classification on ImageNet LT (Sampling Factor Evaluation)
52.9Top-1 Accuracy (s=1.0)Ours(Sum)
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Ours(Sum)Backbone=ResNet-50, Evaluation Protocol=Linear Evaluation2025.09 | 52.9 | 50.25 | 49.04 | 45.51 | 39.68 | |
| MoCoV2Backbone=ResNet-50, Evaluation Protocol=Linear Evaluation2025.09 | 49.5 | 43.2 | 39.5 | 36.6 | 30.5 | |
| SimCLRBackbone=ResNet-50, Evaluation Protocol=Linear Evaluation2025.09 | 46.81 | — | — | — | — | |
| SDCLRBackbone=ResNet-50, Evaluation Protocol=Linear Evaluation2025.09 | 46.62 | — | — | — | — | |
| ByolBackbone=ResNet-50, Evaluation Protocol=Linear Evaluation2025.09 | 43.16 | — | — | — | — | |
| VicRegBackbone=ResNet-50, Evaluation Protocol=Linear Evaluation2025.09 | 38.24 | — | — | — | — |