Fetching the paper…

Self-supervised pre-training and contrastive representation learning for multiple-choice video QA · Around