LMPT: Prompt Tuning with Class-Specific Embedding Loss for Long-tailed Multi-Label Visual Recognition

Open paper
Benchmark Model Rank Results
long-tail-learning-on-coco-mltLMPT(ViT-B/16)#1Average mAP: 66.19
long-tail-learning-on-coco-mltLMPT(ResNet-50)#3Average mAP: 58.97
long-tail-learning-on-voc-mltLMPT(ViT-B/16)#1Average mAP: 87.88
long-tail-learning-on-voc-mltLMPT(ResNet-50)#3Average mAP: 85.44

Rank counts only results with a code link.