You Only Need Less Attention at Each Stage in Vision Transformers
Open paper
Benchmark
Model
Rank
Results
image-classification-on-imagenet
LaViT-B
–
Top 1 Accuracy: 83.1%
object-detection-on-coco-val2017
LaViT-S
–
box AP: 48.9
semantic-segmentation-on-ade20k
LaViT-S
–
Validation mIoU: 47.2
Rank counts only results with a code link.