This is your work, valued
Scale-RAE. Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders
