Image Classification on CIFAR-100-LTN Symmetric Noise IF=10, NR=40% (test)
81.5AccuracyCLIP+CARE
Evaluation results
| Method | Links | |
|---|---|---|
| CLIP+CAREBackbone=CLIP, Noise type=Symmetric, Imbalance Factor (IF)=10, Noise Rate (NR)=40%2026.05 | 81.5 | |
| CLIP+LA w. CAREFoundation Model=CLIP2026.05 | 81.5 | |
| CLIP+RLA*Foundation Model=CLIP2026.05 | 80.9 | |
| CLIP+LAFoundation Model=CLIP2026.05 | 80.5 | |
| CLIP+CE w. CAREFoundation Model=CLIP2026.05 | 80.3 | |
| CLIP+IBCBackbone=CLIP, Noise type=Symmetric, Imbalance Factor (IF)=10, Noise Rate (NR)=40%2026.05 | 79.3 | |
| CLIP+CEFoundation Model=CLIP2026.05 | 78.8 | |
| CLIPProtocol=zero-shot2026.05 | 64.4 | |
| ECBS2026.05 | 56.7 | |
| TABASCO2026.05 | 56.5 | |
| ULC2026.05 | 54.9 | |
| DivideMix2026.05 | 54.7 | |
| UNICON2026.05 | 52.3 | |
| RoLT2026.05 | 42 | |
| HAR2026.05 | 38.2 | |
| CE2026.05 | 34.5 | |
| IB2026.05 | 32.4 | |
| MW-Net2026.05 | 32 | |
| LDAM2026.05 | 31.3 | |
| LA2026.05 | 29.1 |