On the Efficacy of Co-Attention Transformer Layers in Visual Question Answering

ViT
Re-assign community