Parameterization of Cross-Token Relations with Relative Positional Encoding for Vision MLP

Open in new window