Model-Agnostic Human Preference Inversion in Diffusion Models

arXiv:2404.00879v1 Announce Type: new
Abstract: Efficient text-to-image generation remains a challenging task due to the high computational costs associated with the multi-step sampling in diffusion models. Although distillation of pre-trained diffusion models has been successful in reducing sampling steps, low-step image generation often falls short in terms of quality. In this study, we propose a novel sampling design to achieve high-quality one-step image generation aligning with human preferences, particularly focusing on exploring the impact of the prior noise distribution. Our …

abstract arxiv computational costs cs.cv diffusion diffusion models distillation human image image generation low model-agnostic quality sampling study terms text text-to-image type