Our paper “Understanding Cross-modal Interactions in V&L Models that Generate Scene Descriptions” , accepted @UM-IoS Workshop EMNLP 2022, 🇦🇪 🎉