Image Classification on CIFAR-100 LT-100 long-tailed (test)
70.59AccuracyGRAIN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GRAINBackbone=ViT-base, Fine-tuning strategy=FFT + GRAIN2026.06 | 70.59 | 0.38 | |
| ViT (FFT)Backbone=ViT-base, Fine-tuning strategy=FFT2026.06 | 69.55 | 2.71 | |
| PCGradBackbone=ViT-base, Fine-tuning strategy=FFT + PCGrad2026.06 | 68.13 | 1.47 | |
| NoisyTuneBackbone=ViT-base, Fine-tuning strategy=FFT + NoisyTune2026.06 | 68.05 | 3.75 | |
| FocalLossBackbone=ViT-base, Fine-tuning strategy=FFT + FocalLoss2026.06 | 64.68 | 10.7 | |
| SAMBackbone=ViT-base, Fine-tuning strategy=FFT + SAM2026.06 | 63.55 | 11.08 | |
| SWABackbone=ViT-base, Fine-tuning strategy=FFT + SWA2026.06 | 51.14 | 9.6 |