Login / Signup

GvT: A Graph-based Vision Transformer with Talking-Heads Utilizing Sparsity, Trained from Scratch on Small Datasets.

Dongjing ShanGuiqiang Chen
Published in: CoRR (2024)
Keyphrases