Multi-class classification on CIFAR100 LT imbalance factor 0.1 (test)
58.9Top-1 AccuracyRefCal
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=LS Szegedy et al. (2015)2026.05 | 58.9 | 96.65 | 8.76 | 0.4 | 0.47 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=MbLS Liu et al. (2022)2026.05 | 58.79 | 96.62 | 8.6 | 0.41 | 0.47 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=FL + MDCA Hebbalaguppe et al. (2022b)2026.05 | 58.7 | 96.23 | 10.81 | 0.45 | 0.42 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=NLL (CE)2026.05 | 58.46 | 96.22 | 13.79 | 0.47 | 0.44 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=CE + TS Guo et al. (2017)2026.05 | 58.46 | 96.85 | 7.62 | 0.41 | 0.46 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=CRL Moon et al. (2020)2026.05 | 58.46 | 96.22 | 13.8 | 0.47 | 0.44 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=AdaFocal Ghosh et al. (2022)2026.05 | 58.05 | 96.62 | 8.41 | 0.45 | 0.44 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=MixUp Thulasidasan et al. (2019)2026.05 | 56.35 | 96.37 | 18.84 | 0.69 | 0.72 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=MMCE Kumar et al. (2018)2026.05 | 55.9 | 95.03 | 7.91 | 0.54 | 0.56 | |
| RefCalRefinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-50, Baseline Comparison=LogitNorm Wei et al. (2022)2026.05 | 55.89 | 96.3 | 25.97 | 0.68 | 0.97 | |
| LogitNorm Wei et al. (2022)Refinement Method (Ref.)=true, Calibration Method (Cal.)=false, Backbone=ResNet-502026.05 | 52.96 | 94.74 | 49.84 | 0.13 | 1.3 | |
| MixUp Thulasidasan et al. (2019)Refinement Method (Ref.)=false, Calibration Method (Cal.)=true, Backbone=ResNet-502026.05 | 52.9 | 95.82 | 6.1 | 0.58 | 0.54 | |
| MMCE Kumar et al. (2018)Refinement Method (Ref.)=false, Calibration Method (Cal.)=true, Backbone=ResNet-502026.05 | 49.11 | 94.4 | 25.9 | 0.65 | 0.5 | |
| MbLS Liu et al. (2022)Refinement Method (Ref.)=false, Calibration Method (Cal.)=true, Backbone=ResNet-502026.05 | 48.1 | 93 | 8.36 | 0.45 | 0.47 | |
| LS Szegedy et al. (2015)Refinement Method (Ref.)=false, Calibration Method (Cal.)=true, Backbone=ResNet-502026.05 | 47.84 | 92.11 | 9.21 | 0.44 | 0.55 | |
| AdaFocal Ghosh et al. (2022)Refinement Method (Ref.)=false, Calibration Method (Cal.)=true, Backbone=ResNet-502026.05 | 47.68 | 95.49 | 29 | 0.72 | 0.55 | |
| NLL (CE)Refinement Method (Ref.)=true, Calibration Method (Cal.)=false, Backbone=ResNet-502026.05 | 47.57 | 94 | 21.7 | 0.61 | 0.52 | |
| CRL Moon et al. (2020)Refinement Method (Ref.)=true, Calibration Method (Cal.)=false, Backbone=ResNet-502026.05 | 46.27 | 93.7 | 22.03 | 0.63 | 0.54 | |
| FL + MDCA Hebbalaguppe et al. (2022b)Refinement Method (Ref.)=false, Calibration Method (Cal.)=true, Backbone=ResNet-502026.05 | 46.17 | 94.16 | 11.32 | 0.52 | 0.48 | |
| CE + TS Guo et al. (2017)Refinement Method (Ref.)=true, Calibration Method (Cal.)=true, Backbone=ResNet-502026.05 | 45.35 | 94.4 | 28.25 | 0.71 | 0.57 |