BtR

Reported on 6 benchmarks across 2 tasks · 1 paper

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

Computer Vision5 results

Image Classificationonmini WebVision 1.0
ImageNet Top-1 Accuracy· 2022-10-17
75.96
best: 82.56 (LRA-diffusion (CLIP ViT))
Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels arXiv:2210.08826
Image Classificationonmini WebVision 1.0
ImageNet Top-5 Accuracy· 2022-10-17
92.2
best: 97.24 (RTE (Inception-ResNet-v2))
Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels arXiv:2210.08826
Image Classificationonmini WebVision 1.0
Top-1 Accuracy· 2022-10-17
80.88
best: 84.16 (LRA-diffusion (CLIP ViT))
Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels arXiv:2210.08826
Image Classificationonmini WebVision 1.0
Top-5 Accuracy· 2022-10-17
92.76
best: 94.84 (PSSCL (130 epochs))
Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels arXiv:2210.08826
Image ClassificationonANIMAL
Accuracy· 2022-10-17
88.5
best: 89 (Jigsaw-ViT)
Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels arXiv:2210.08826

Medical1 result

Document Text ClassificationonANIMAL
Accuracy· 2022-10-17
88.5
best: 89 (Jigsaw-ViT)
Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels arXiv:2210.08826