OneSparse: A Unified Framework for Sparse Activation Layers in Vision Models
Open paper
Benchmark
Model
Rank
Results
image-classification-on-imagenet
Nexus (ConvNeXt-B)
–
Top 1 Accuracy: 84.5%
object-detection-on-coco-val2017
Nexus (ConvNeXt-S)
–
box AP: 48.4
Rank counts only results with a code link.