Image Classification on CIFAR-10 LT-100 long-tailed (test)
89.83AccuracyGRAIN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GRAINBackbone=ViT-base, Fine-tuning strategy=FFT + GRAIN2026.06 | 89.83 | 0.88 | |
| FocalLossBackbone=ViT-base, Fine-tuning strategy=FFT + FocalLoss2026.06 | 89.32 | 0.68 | |
| ViT (FFT)Backbone=ViT-base, Fine-tuning strategy=FFT2026.06 | 88.47 | 1.77 | |
| SAMBackbone=ViT-base, Fine-tuning strategy=FFT + SAM2026.06 | 88.07 | 9.41 | |
| NoisyTuneBackbone=ViT-base, Fine-tuning strategy=FFT + NoisyTune2026.06 | 87.88 | 1.67 | |
| PCGradBackbone=ViT-base, Fine-tuning strategy=FFT + PCGrad2026.06 | 86.63 | 2.44 | |
| SWABackbone=ViT-base, Fine-tuning strategy=FFT + SWA2026.06 | 39.29 | 4.29 |