Search Results for author:
Found 2 papers, 0 papers with code
Token Recycling for Efficient Sequential Inference with Vision Transformers
To reduce this computational inefficiency, we introduce the TOken REcycling (TORE) modification for the ViT inference, which can be used with any architecture.
Learning Nuclei Representations with Masked Image Modelling
Inspired by Bidirectional Encoder representation from Image Transformers (BEiT), we split the images into smaller patches and generate corresponding discrete visual tokens.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
