Long-tailed Image Classification on CIFAR10-LT IF 50 (test)
96.97AccuracyGRAIN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GRAINBackbone=ViT-base, Fine-tuning strategy=FFT + GRAIN2026.06 | 96.97 | 0.13 | |
| PCGradBackbone=ViT-base, Fine-tuning strategy=FFT + PCGrad2026.06 | 96.85 | 0.19 | |
| ViT (FFT)Backbone=ViT-base, Fine-tuning strategy=FFT2026.06 | 96.46 | 0.44 | |
| FocalLossBackbone=ViT-base, Fine-tuning strategy=FFT + FocalLoss2026.06 | 95.29 | 3.58 | |
| SAMBackbone=ViT-base, Fine-tuning strategy=FFT + SAM2026.06 | 92.15 | 11.55 | |
| SWABackbone=ViT-base, Fine-tuning strategy=FFT + SWA2026.06 | 90.58 | 5.31 | |
| NoisyTuneBackbone=ViT-base, Fine-tuning strategy=FFT + NoisyTune2026.06 | 88.94 | 9.39 | |
| NCLAggregation=ensemble2022.03 | 87.3 | — | |
| NCLAggregation=single expert2022.03 | 86.8 | — | |
| BSCEVariant=baseline2022.03 | 85.8 | — | |
| MISLASReference=CVPR'212022.03 | 85.7 | — | |
| Hybrid-SCReference=CVPR'212022.03 | 85.4 | — | |
| ACEReference=ICCV'212022.03 | 84.9 | — | |
| FBLBackbone Net=ResNet-32, IF=502023.05 | 84.3 | — | |
| CAMReference=AAAI'212022.03 | 83.6 | — | |
| LDAM+DAPReference=CVPR'202022.03 | 82.2 | — | |
| BBNReference=CVPR'202022.03 | 82.2 | — | |
| BBNBackbone Net=ResNet-32, IF=502023.05 | 81.18 | — | |
| LDAM-DRWBackbone Net=ResNet-32, IF=502023.05 | 81.03 | — | |
| CB Focal lossReference=CVPR'192022.03 | 79.3 | — | |
| LDAM+DRWReference=NeurIPS'192022.03 | 79.3 | — | |
| CE lossBackbone Net=ResNet-32, IF=502023.05 | 75.31 | — |