- Results on UDM10, REDS30 and MVSR4x. (Tab. 2 of the main paper)
LSGQuant is a layer-sensitivity guided quantizing approach for one-step diffusion-based real-world VSR.
LSGQuant is a layer-sensitivity guided quantizing approach for one-step diffusion-based real-world VSR.
One-Step Diffusion Models have demonstrated promising capability and fast inference in video super-resolution (VSR) for real-world. Nevertheless, the substantial model size and high computational cost of Diffusion Transformers (DiTs) limit downstream applications. While low-bit quantization is a common approach for model compression, the effectiveness of quantized models is challenged by the high dynamic range of input latent and diverse layer behaviors. To deal with these challenges, we introduce LSGQuant, a layer-sensitivity guided quantizing approach for one-step diffusion-based real-world VSR. Our method incorporates a Dynamic Range Adaptive Quantizer (DRAQ) to fit video token activations. Furthermore, we estimate layer sensitivity and implement a Variance-Oriented Layer Training Strategy (VOLTS) by analyzing layer-wise statistics in calibration. We also introduce Quantization-Aware Optimization (QAO) to jointly refine the quantized branch and a retained high-precision branch. Extensive experiments demonstrate that our method has nearly performance to origin model with full-precision and significantly exceeds existing quantization techniques.
Overview of our LSGQuant. Firstly, we collect per-channel layer statistics by a single inference process. Next, we calculate layer sensitivity and estimate its importance to the final video output. Finally, we allocate training iterations by estimated sensitivities.
@article{wu2026lsgquant,
title = {LSGQuant: Layer-Sensitivity Guided Quantization for One-Step Diffusion Real-World Video Super-Resolution},
author = {Wu, Tianxing and Chen, Zheng and Xu, Cirou and Chai, Bowen and Guo, Yong and Liu, Yutong and Kong, Linghe and Zhang, Yulun},
journal = {arXiv preprint arXiv:2602.03182},
year = {2026}
}