PixelDiT: Pixel Diffusion Transformers for Image Generationsion Transformers for Image Generation

PixelGen: Improving Pixel Diffusion with Perceptual Supervision

DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation

DiP: Taming Diffusion Models in Pixel Space

L2P: UNLOCKING LATENT POTENTIAL FOR PIXEL GENERATION

Pixel-Space Diffusion Transformers

PixelREPA

An Empirical Study of Training Pixel-Space Text-to-Image Diffusion Models

SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture