Transformer-based Cross-Modal Recipe Embeddings with Large Batch Training

ViT
Re-assign community