RowFormer: Multiple Class-Token-Based Vision Transformer for 2D Context-Aware Attention.

Mohamed Dahmane, Mamadou Dian Bah, Rajjeshwar Ganguly

Browse the full ACIVS paper archive.