Fetching the paper…

Contrastive Representation Learning with Trainable Augmentation Channel · Around