Fetching the paper…

AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognition · Around