Looped Diffusion Transformer
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 parameter count fixed. This looped computation enables iterative refinement of internal representations without explicit reasoning tokens. However, naive looping fails to consistently improve image quality. We trace this problem to weak supervision across intermediate loops and unregulated attention updates that progressively erode local information. To overcome these challenge...
| Authors | Yong Xien Chng, Tianyi Chen, Wenwen Tong et al. |
|---|---|
| Primary category | cs.CV |
| Published | 30 September 2026 |
| Last updated at source | 30 September 2026 |
Connected in the knowledge graph
- Jinyan0924/flux3-agibot-loopedModel · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.