Paper Detail

PixelDiT2: Representation-Grounded Pixel Diffusion Transformers

Yongsheng Yu, Wei Xiong, Yichen Sheng, Shiqiu Liu, Jiebo Luo

arxiv Score 7.3

Published 2026-09-21 · First seen 2026-09-22

General AI

Abstract

Recent advances in pixel-space diffusion models have narrowed the image quality gap with latent-space diffusion, but still converge more slowly and lag behind in final image quality. We argue that a key reason is the lack of an explicit representation prior: unlike latent diffusion, which usually denoises in a compact and structured latent space, pixel diffusion needs to learn denoising-friendly representations and pixel generation simultaneously from raw RGB space. To address this problem, we propose PixelDiT2, an end-to-end pixel-space diffusion model designed to decouple representation learning from pixel generation without introducing an autoencoder or latent reconstruction bottleneck. We propose representation grounding that uses a frozen pretrained vision foundation model to provide explicit per-patch representation guidance throughout denoising, allowing the pixel diffusion transformer to focus more on pixel generation. On ImageNet-256x256, PixelDiT2 achieves an FID of 1.46 after 600 epochs; at 512x512 resolution, PixelDiT2 achieves an FID of 1.48 after 680 epochs.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
later
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{yu2026pixeldit2,
  title = {PixelDiT2: Representation-Grounded Pixel Diffusion Transformers},
  author = {Yongsheng Yu and Wei Xiong and Yichen Sheng and Shiqiu Liu and Jiebo Luo},
  year = {2026},
  abstract = {Recent advances in pixel-space diffusion models have narrowed the image quality gap with latent-space diffusion, but still converge more slowly and lag behind in final image quality. We argue that a key reason is the lack of an explicit representation prior: unlike latent diffusion, which usually denoises in a compact and structured latent space, pixel diffusion needs to learn denoising-friendly representations and pixel generation simultaneously from raw RGB space. To address this problem, we p},
  url = {https://arxiv.org/abs/2609.24919},
  keywords = {cs.CV},
  eprint = {2609.24919},
  archiveprefix = {arXiv},
}

Metadata

{}