Deep Reward Supervisions for Tuning Text-to-Image Diffusion Models

arXiv:2405.00760v1 Announce Type: new
Abstract: Optimizing a text-to-image diffusion model with a given reward function is an important but underexplored research area. In this study, we propose Deep Reward Tuning (DRTune), an algorithm that directly supervises the final output image of a text-to-image diffusion model and back-propagates through the iterative sampling process to the input noise. We find that training earlier steps in the sampling process is crucial for low-level rewards, and deep supervision can be achieved efficiently and effectively …

abstract algorithm arxiv cs.ai cs.cv diffusion diffusion model diffusion models function image image diffusion iterative process research sampling study text text-to-image through type