| document-layout-analysis-on-publaynet-val | BEiT-B | #7 | Overall: 0.931Text: 0.934Title: 0.866List: 0.924Table: 0.973… |
| self-supervised-image-classification-on-imagenet-finetuned | BEiT-L (ViT) | #14 | Top 1 Accuracy: 86.3%Number of Params: 307M |
| self-supervised-image-classification-on-imagenet-finetuned | BEiT-B (ViT) | #29 | Top 1 Accuracy: 84.6%Number of Params: 86M |
| semantic-segmentation-on-ade20k-val | BEiT-L (ViT+UperNet, ImageNet-22k pretrain) | #21 | mIoU: 57.0 |
| image-classification-on-omnibenchmark | BeiT | #22 | Average Top-1 Accuracy: 30.1 |
| document-image-classification-on-rvl-cdip | BEiT-B | #24 | Accuracy: 91.09%Parameters: 87M |
| image-classification-on-cifar-100 | BEiT (extra data) | #25 | Percentage correct: 91.8 |
| semantic-segmentation-on-ade20k | BEiT-L (ViT+UperNet) | #36 | Validation mIoU: 57.0 |
| image-classification-on-imagenet | BEiT-L (ViT; ImageNet-22K pretrain) | #40 | Top 1 Accuracy: 88.60%Number of params: 331M |
| image-classification-on-imagenet | BEiT-L (ViT; ImageNet 1k pretrain) | #168 | Top 1 Accuracy: 86.3%Number of params: 86M |