TL;DR
DuoDiT is introduced, a parameter-efficient dual-stream architecture that keeps a pre-trained DiT-XL/2 backbone frozen and adds a trainable fine-grained Second Stream, which obtains the lowest KID and LPIPS-Mean while maintaining competitive FID.
PixRestore is presented, a VAE-free pixel-space Diffusion Transformer (DiT) for UIR, where the diffusion backbone is trained entirely from scratch, without relying on T2I pretraining.
Generating images beyond the training resolution of Diffusion Transformers (DiTs) remains challenging, as the number of tokens grows rapidly with resolution, full attention becomes expensive, and outputs often suffer from structural distortions and degraded detail. We identify uniform full attention as the common sourc...
Improving text-to-image models has traditionally relied on increasing model size or the number of denoising steps. In this work, we explore an alternative way to scale computation by repeatedly running shared Transformer blocks within each denoising step, effectively increasing computational depth while keeping the par...
High-resolution remote sensing images present considerable challenges for semantic segmentation due to their complex object structures and extensive spatial distribution. Effective segmentation requires capturing fine-grained local details while simultaneously modeling long-range dependencies. Convolutional Neural Netw...
Video diffusion models (VDMs) have achieved impressive progress in text-to-video generation, but their high memory and computational costs hinder practical deployment. Quantization-aware training (QAT) is an effective solution for compressing and accelerating advanced generative models without runtime overhead at infer...
Post-hoc linearization is an attractive way to accelerate pretrained Video Diffusion Transformers (Video DiTs), yet existing methods largely treat the quality loss induced by attention replacement as a recovery problem. We show that a more fundamental bottleneck lies upstream: \emph{initialization}. Random initializati...
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
