Timestep-Aware Correction for Quantized Diffusion Models

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Yao, Yuzhe, Tian, Feng, Chen, Jun, Lin, Haonan, Dai, Guang, Liu, Yong, Wang, Jingdong
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910514259427328
author Yao, Yuzhe
Tian, Feng
Chen, Jun
Lin, Haonan
Dai, Guang
Liu, Yong
Wang, Jingdong
author_facet Yao, Yuzhe
Tian, Feng
Chen, Jun
Lin, Haonan
Dai, Guang
Liu, Yong
Wang, Jingdong
contents Diffusion models have marked a significant breakthrough in the synthesis of semantically coherent images. However, their extensive noise estimation networks and the iterative generation process limit their wider application, particularly on resource-constrained platforms like mobile devices. Existing post-training quantization (PTQ) methods have managed to compress diffusion models to low precision. Nevertheless, due to the iterative nature of diffusion models, quantization errors tend to accumulate throughout the generation process. This accumulation of error becomes particularly problematic in low-precision scenarios, leading to significant distortions in the generated images. We attribute this accumulation issue to two main causes: error propagation and exposure bias. To address these problems, we propose a timestep-aware correction method for quantized diffusion model, which dynamically corrects the quantization error. By leveraging the proposed method in low-precision diffusion models, substantial enhancement of output quality could be achieved with only negligible computation overhead. Extensive experiments underscore our method's effectiveness and generalizability. By employing the proposed correction strategy, we achieve state-of-the-art (SOTA) results on low-precision models.
format Preprint
id arxiv_https___arxiv_org_abs_2407_03917
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Timestep-Aware Correction for Quantized Diffusion Models
Yao, Yuzhe
Tian, Feng
Chen, Jun
Lin, Haonan
Dai, Guang
Liu, Yong
Wang, Jingdong
Computer Vision and Pattern Recognition
Diffusion models have marked a significant breakthrough in the synthesis of semantically coherent images. However, their extensive noise estimation networks and the iterative generation process limit their wider application, particularly on resource-constrained platforms like mobile devices. Existing post-training quantization (PTQ) methods have managed to compress diffusion models to low precision. Nevertheless, due to the iterative nature of diffusion models, quantization errors tend to accumulate throughout the generation process. This accumulation of error becomes particularly problematic in low-precision scenarios, leading to significant distortions in the generated images. We attribute this accumulation issue to two main causes: error propagation and exposure bias. To address these problems, we propose a timestep-aware correction method for quantized diffusion model, which dynamically corrects the quantization error. By leveraging the proposed method in low-precision diffusion models, substantial enhancement of output quality could be achieved with only negligible computation overhead. Extensive experiments underscore our method's effectiveness and generalizability. By employing the proposed correction strategy, we achieve state-of-the-art (SOTA) results on low-precision models.
title Timestep-Aware Correction for Quantized Diffusion Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.03917