Image Classification on CIFAR-100 LT-50 long-tailed (test)
88.87AccuracyGRAIN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GRAINBackbone=ViT-base, Fine-tuning strategy=FFT + GRAIN2026.06 | 88.87 | 0.09 | |
| NoisyTuneBackbone=ViT-base, Fine-tuning strategy=FFT + NoisyTune2026.06 | 87.89 | 0.24 | |
| SAMBackbone=ViT-base, Fine-tuning strategy=FFT + SAM2026.06 | 87.81 | 0.91 | |
| FocalLossBackbone=ViT-base, Fine-tuning strategy=FFT + FocalLoss2026.06 | 87.2 | 2.21 | |
| PCGradBackbone=ViT-base, Fine-tuning strategy=FFT + PCGrad2026.06 | 87.02 | 0.37 | |
| ViT (FFT)Backbone=ViT-base, Fine-tuning strategy=FFT2026.06 | 85.53 | 0.48 | |
| SWABackbone=ViT-base, Fine-tuning strategy=FFT + SWA2026.06 | 66.96 | 13.46 |