Long-Tailed Recognition by Mutual Information Maximization between Latent Features and Ground-Truth Labels

Open paper
Benchmark Model Rank Results
long-tail-learning-on-cifar-100-lt-r-10GML (ResNet-32)#10Error Rate: 33.0
long-tail-learning-on-cifar-100-lt-r-100GML (ResNet-32)#13Error Rate: 46.0
long-tail-learning-on-cifar-100-lt-r-50GML (ResNet-32)#11Error Rate: 41.9
long-tail-learning-on-imagenet-ltGML (ResNeXt-50)#18Top-1 Accuracy: 58.8
long-tail-learning-on-inaturalist-2018GML (ViT-B-16)#3Top-1 Accuracy: 82.1%
long-tail-learning-on-inaturalist-2018GML (ResNet-50)#17Top-1 Accuracy: 74.5%

Rank counts only results with a code link.