ConvMLP-L

Reported on 5 benchmarks across 3 tasks · 1 paper

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

Computer Vision3 results

Image ClassificationonCIFAR-10
Percentage correct· uses extra data· 2021-09-09
98.6
best: 99.5 (ViT-H/14)
ConvMLP: Hierarchical Convolutional MLPs for Vision arXiv:2109.04454
Image ClassificationonFlowers-102
Accuracy· uses extra data· 2021-09-09
99.5
best: 99.76 (CCT-14/7x2)
ConvMLP: Hierarchical Convolutional MLPs for Vision arXiv:2109.04454
Image ClassificationonCIFAR-100
Percentage correct· uses extra data· 2021-09-09
88.6
best: 96.08 (EffNet-L2 (SAM))
ConvMLP: Hierarchical Convolutional MLPs for Vision arXiv:2109.04454

Medical1 result

Semantic SegmentationonADE20K
Validation mIoU· 2021-09-09
40
best: 63.6 (ViT-P (InternImage-H))
ConvMLP: Hierarchical Convolutional MLPs for Vision arXiv:2109.04454

Audio1 result

10-shot image generationonADE20K
Validation mIoU· 2021-09-09
40
best: 63.6 (ViT-P (InternImage-H))
ConvMLP: Hierarchical Convolutional MLPs for Vision arXiv:2109.04454