Metraipolis › archive › dc_167f4244ff1a
Training-Free Hidden-State Refinement for Flow-Matching Image Generators — arXiv:2608.29160
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.29160
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:47:31.167Z
- SHA-256 of the archived text
4923ddeec4d9c1c81bbf1a5a5cc3d67e61ae7d78475fbbe2871698c2a0c9f86c- Stable URL of this page
https://metraipolis.com/documento/dc_167f4244ff1a
How to cite
Training-Free Hidden-State Refinement for Flow-Matching Image Generators — arXiv:2608.29160 — arXiv (en), https://arxiv.org/abs/2608.29160 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_167f4244ff1a (Metraipolis archive dc_167f4244ff1a, sha256 4923ddeec4d9c1c81bbf1a5a5cc3d67e61ae7d78475fbbe2871698c2a0c9f86c)
BibTeX
@misc{dc167f4244ff1a,
title = {Training-Free Hidden-State Refinement for Flow-Matching Image Generators — arXiv:2608.29160},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.29160},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_167f4244ff1a; sha256 4923ddeec4d9c1c81bbf1a5a5cc3d67e61ae7d78475fbbe2871698c2a0c9f86c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.29160. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We aim to improve frozen flow-matching image generators by adding inference computation inside the denoiser, without changing model weights or the outer sampler. Existing generators usually spend extra test-time computation by increasing the number of sampling steps, which repeatedly evaluates the entire denoiser and couples quality gains to sampler cost. A key challenge is how to use extra computation inside a frozen transformer denoiser: the method must decide which tokens, layers, and sampling times receive repeated updates while preserving the original generation pipeline. We introduce a training-free looping framework that repeatedly applies selected transformer layers inside each denoising call. Dense and Sparse Token Loop vary the token scope; Sampling-Progress Gating and the loop layer range specify when and where looping is active; loop count and strength control the repeated updates; and Loop Guidance combines ordinary and looped vector-field predictions. Across two Scale-RAE model scales, loop variants improve primary and auxiliary quality metrics with competitive quality--efficiency trade-offs. Loop Guidance further improves both primary metrics across all three tested models; on Scale-RAE DiT2.4B, it raises GenEval from 0.4471 to 0.5691 and DPG-Bench from 0.7656 to 0.8053. Code will be released. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.29160 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T07:28:26.723Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).