Explicit Cross-Modal Representation Learning for Visual Commonsense Reasoning | IEEE Journals & Magazine | IEEE Xplore