Learning with Neighbor Consistency for Noisy Labels

Open paper
Benchmark Model Rank Results
image-classification-on-mini-webvision-1-0NCR+Mixup+DA (ResNet-50)#5Top-1 Accuracy: 80.5
image-classification-on-mini-webvision-1-0NCR+Mixup (ResNet-50)#13Top-1 Accuracy: 79.4
image-classification-on-mini-webvision-1-0NCR (ResNet-50)#25Top-1 Accuracy: 77.1
image-classification-on-red-miniimagenet-20NCR (ResNet-18)#1Accuracy: 69.0
image-classification-on-red-miniimagenet-40NCR (ResNet-18)#1Accuracy: 64.6
image-classification-on-red-miniimagenet-80NCR (ResNet-18)#1Accuracy: 51.2
image-classification-on-webvision-1000NCR+Mixup+DA (ResNet-50)#3Top-1 Accuracy: 76.8
image-classification-on-webvision-1000NCR (ResNet-50)#5Top-1 Accuracy: 75.7%
learning-with-noisy-labels-on-redNCR (ResNet-18)#1Test Accuracy: 69.0
learning-with-noisy-labels-on-red-1NCR (ResNet-18)#1Test Accuracy: 64.6
learning-with-noisy-labels-on-red-3NCR (ResNet-18)#1Test Accuracy: 51.2

Rank counts only results with a code link.