Targeted Supervised Contrastive Learning for Long-Tailed Recognition

Open paper
Benchmark Model Rank Results
long-tail-learning-on-cifar-10-lt-r-10TSC#26Error Rate: 11.3
long-tail-learning-on-cifar-10-lt-r-100TSC(ResNet-32)#22Error Rate: 21.3
long-tail-learning-on-cifar-100-lt-r-100TSC(ResNet-32)#47Error Rate: 56.2
long-tail-learning-on-imagenet-ltTSC(ResNet-50)#47Top-1 Accuracy: 52.4
long-tail-learning-on-inaturalist-2018TSC(ResNet-50)#34Top-1 Accuracy: 69.7%

Rank counts only results with a code link.