Leveraging Early-Stage Robustness in Diffusion Models for Efficient and High-Quality Image Synthesis

Published: 21 Sept 2023, Last Modified: 21 Dec 2023NeurIPS 2023 posterEveryoneRevisionsBibTeX
Keywords: diffusion models, post-training quantization
Abstract: While diffusion models have demonstrated exceptional image generation capabilities, the iterative noise estimation process required for these models is compute-intensive and their practical implementation is limited by slow sampling speeds. In this paper, we propose a novel approach to speed up the noise estimation network by leveraging the robustness of early-stage diffusion models. Our findings indicate that inaccurate computation during the early-stage of the reverse diffusion process has minimal impact on the quality of generated images, as this stage primarily outlines the image while later stages handle the finer details that require more sensitive information. To improve computational efficiency, we combine our findings with post-training quantization (PTQ) to introduce a method that utilizes low-bit activation for the early reverse diffusion process while maintaining high-bit activation for the later stages. Experimental results show that the proposed method can accelerate the early-stage computation without sacrificing the quality of the generated images.
Supplementary Material: zip
Submission Number: 5879