VARestorer: One-Step VAR Distillation for Real-World Image Super-Resolution

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhu, Yixuan, Ma, Shilin, Wang, Haolin, Li, Ao, Jing, Yanzhe, Tang, Yansong, Chen, Lei, Lu, Jiwen, Zhou, Jie
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908989297524736
author Zhu, Yixuan
Ma, Shilin
Wang, Haolin
Li, Ao
Jing, Yanzhe
Tang, Yansong
Chen, Lei
Lu, Jiwen
Zhou, Jie
author_facet Zhu, Yixuan
Ma, Shilin
Wang, Haolin
Li, Ao
Jing, Yanzhe
Tang, Yansong
Chen, Lei
Lu, Jiwen
Zhou, Jie
contents Recent advancements in visual autoregressive models (VAR) have demonstrated their effectiveness in image generation, highlighting their potential for real-world image super-resolution (Real-ISR). However, adapting VAR for ISR presents critical challenges. The next-scale prediction mechanism, constrained by causal attention, fails to fully exploit global low-quality (LQ) context, resulting in blurry and inconsistent high-quality (HQ) outputs. Additionally, error accumulation in the iterative prediction severely degrades coherence in ISR task. To address these issues, we propose VARestorer, a simple yet effective distillation framework that transforms a pre-trained text-to-image VAR model into a one-step ISR model. By leveraging distribution matching, our method eliminates the need for iterative refinement, significantly reducing error propagation and inference time. Furthermore, we introduce pyramid image conditioning with cross-scale attention, which enables bidirectional scale-wise interactions and fully utilizes the input image information while adapting to the autoregressive mechanism. This prevents later LQ tokens from being overlooked in the transformer. By fine-tuning only 1.2\% of the model parameters through parameter-efficient adapters, our method maintains the expressive power of the original VAR model while significantly enhancing efficiency. Extensive experiments show that VARestorer achieves state-of-the-art performance with 72.32 MUSIQ and 0.7669 CLIPIQA on DIV2K dataset, while accelerating inference by 10 times compared to conventional VAR inference.
format Preprint
id arxiv_https___arxiv_org_abs_2604_21450
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle VARestorer: One-Step VAR Distillation for Real-World Image Super-Resolution
Zhu, Yixuan
Ma, Shilin
Wang, Haolin
Li, Ao
Jing, Yanzhe
Tang, Yansong
Chen, Lei
Lu, Jiwen
Zhou, Jie
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Recent advancements in visual autoregressive models (VAR) have demonstrated their effectiveness in image generation, highlighting their potential for real-world image super-resolution (Real-ISR). However, adapting VAR for ISR presents critical challenges. The next-scale prediction mechanism, constrained by causal attention, fails to fully exploit global low-quality (LQ) context, resulting in blurry and inconsistent high-quality (HQ) outputs. Additionally, error accumulation in the iterative prediction severely degrades coherence in ISR task. To address these issues, we propose VARestorer, a simple yet effective distillation framework that transforms a pre-trained text-to-image VAR model into a one-step ISR model. By leveraging distribution matching, our method eliminates the need for iterative refinement, significantly reducing error propagation and inference time. Furthermore, we introduce pyramid image conditioning with cross-scale attention, which enables bidirectional scale-wise interactions and fully utilizes the input image information while adapting to the autoregressive mechanism. This prevents later LQ tokens from being overlooked in the transformer. By fine-tuning only 1.2\% of the model parameters through parameter-efficient adapters, our method maintains the expressive power of the original VAR model while significantly enhancing efficiency. Extensive experiments show that VARestorer achieves state-of-the-art performance with 72.32 MUSIQ and 0.7669 CLIPIQA on DIV2K dataset, while accelerating inference by 10 times compared to conventional VAR inference.
title VARestorer: One-Step VAR Distillation for Real-World Image Super-Resolution
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2604.21450