Sel-CL+ (ResNet-18)

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

Image Classificationonmini WebVision 1.0
ImageNet Top-1 Accuracy· 2022-03-08
76.84
best: 82.56 (LRA-diffusion (CLIP ViT))
Selective-Supervised Contrastive Learning with Noisy Labels arXiv:2203.04181
Image Classificationonmini WebVision 1.0
ImageNet Top-5 Accuracy· 2022-03-08
93.04
best: 97.24 (RTE (Inception-ResNet-v2))
Selective-Supervised Contrastive Learning with Noisy Labels arXiv:2203.04181
Image Classificationonmini WebVision 1.0
Top-1 Accuracy· 2022-03-08
79.96
best: 84.16 (LRA-diffusion (CLIP ViT))
Selective-Supervised Contrastive Learning with Noisy Labels arXiv:2203.04181
Image Classificationonmini WebVision 1.0
Top-5 Accuracy· 2022-03-08
92.64
best: 94.84 (PSSCL (130 epochs))
Selective-Supervised Contrastive Learning with Noisy Labels arXiv:2203.04181