Fetching the paper…

InterBERT: Vision-and-Language Interaction for Multi-modal Pretraining · Around