LookFlow: Training-Free and Efficient High-Resolution Image Synthesis via Dynamic Lookahead Guidance Flow
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (16) · 13800-13808
DOI: 10.1609/aaai.v40i16.38388 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yuan Zhou und 7 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).
Abstract (Originalsprache)
Rectification flow Transformers (RFTs) have shown promising performance in diffusion-based image synthesis but are typically confined to lower-resolution scenarios, limiting their ability to generate high-resolution images. Existing resolution extrapolation approaches often suffer from excessive computational overhead, resulting in prolonged inference times. We propose LookFlow, a training-free high-resolution synthesis framework that accelerates inference while preserving visual quality. Building on pretrained text-to-image RFTs, LookFlow employs a dynamic lookahead guidance flow mechanism to refine high-resolution velocity predictions by leveraging multi-timestep lookahead information extracted from a low-resolution flow. Additionally, reusing temporally similar features across consecutive timesteps drastically reduces computation and significantly decreases inference time overhead. Extensive experiments on COCO demonstrate that LookFlow robustly scales resolutions from 4× to 25×, achieving up to a maximum speedup of 2.01× while maintaining competitive visual fidelity.
Zitieren
APA
Zhou, Y., Zhang, Y., Chang, J., Gu, X., Wang, Y., Ding, K., Yang, G., & Xiang, S. (2026). LookFlow: Training-Free and Efficient High-Resolution Image Synthesis via Dynamic Lookahead Guidance Flow. Proceedings of the AAAI Conference on Artificial Intelligence, 40(16), 13800-13808. https://doi.org/10.1609/aaai.v40i16.38388
Vancouver
Zhou Y, Zhang Y, Chang J, Gu X, Wang Y, Ding K, et al. LookFlow: Training-Free and Efficient High-Resolution Image Synthesis via Dynamic Lookahead Guidance Flow. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(16):13800-13808. doi:10.1609/aaai.v40i16.38388
BibTeX
@article{Zhou2026,
author = {Zhou, Yuan and Zhang, Yan and Chang, Jianlong and Gu, Xin and Wang, Ying and Ding, Kun and Yang, Guangwen and Xiang, Shiming},
title = {LookFlow: Training-Free and Efficient High-Resolution Image Synthesis via Dynamic Lookahead Guidance Flow},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {16},
pages = {13800--13808},
doi = {10.1609/aaai.v40i16.38388}
}
EUVIMED