Paper Detail
Dengyang Jiang, Ruoyi Du, Zhennan Chen, Dongyang Liu, Zanyi Wang, Mingzhe Zheng, Xiangpeng Yang, Huanqia Cai, Aiming Hao, Yuming Jiang, Peng Gao, Harry Yang, Steven Hoi
This paper investigates an increasingly important topic in generative modeling: pixel-space diffusion models. Although numerous studies have explored this topic, most focus on small-scale or class-conditional settings. Consequently, a practical recipe for training pixel-space models that rival or exceed well-established latent-space counterparts remains elusive. Through a comprehensive empirical study, we first observe that direct large-scale pre-training in pixel space converges substantially more slowly than in latent space. This observation motivates a latent-to-pixel strategy that acquires generative priors efficiently in latent space and transitions to pixel space during post-training. We then systematically investigate the key design choices governing this transition, including weight initialization, data composition, prediction target, decoder architecture, and noise schedule, and identify a practical recipe that makes the resulting pixel-space models match or outperform their latent-space counterparts while delivering 3.18 to 4.75 times end-to-end inference speedups. We hope that our findings provide useful empirical insights and practical guidelines for future research on pixel-space generation.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@article{jiang2026empirical,
title = {An Empirical Study of Training Pixel-Space Text-to-Image Diffusion Models},
author = {Dengyang Jiang and Ruoyi Du and Zhennan Chen and Dongyang Liu and Zanyi Wang and Mingzhe Zheng and Xiangpeng Yang and Huanqia Cai and Aiming Hao and Yuming Jiang and Peng Gao and Harry Yang and Steven Hoi},
year = {2026},
abstract = {This paper investigates an increasingly important topic in generative modeling: pixel-space diffusion models. Although numerous studies have explored this topic, most focus on small-scale or class-conditional settings. Consequently, a practical recipe for training pixel-space models that rival or exceed well-established latent-space counterparts remains elusive. Through a comprehensive empirical study, we first observe that direct large-scale pre-training in pixel space converges substantially m},
url = {https://arxiv.org/abs/2608.16887},
keywords = {cs.CV, pixel-space diffusion models, latent-to-pixel strategy, generative priors, weight initialization, prediction target, decoder architecture, noise schedule, huggingface daily},
eprint = {2608.16887},
archiveprefix = {arXiv},
}
{}