Self-Rewarding Large Vision-Language Models for Optimizing Prompts in Text-to-Image Generation
This addresses the challenge of prompt optimization for users of text-to-image models, though it is incremental as it builds on existing methods with a novel self-improvement approach.
The paper tackles the problem of crafting sophisticated prompts for text-to-image generation by proposing a self-rewarding framework that uses large vision-language models to rewrite prompts and assess image quality, outperforming competitors on two datasets.
Text-to-image models are powerful for producing high-quality images based on given text prompts, but crafting these prompts often requires specialized vocabulary. To address this, existing methods train rewriting models with supervision from large amounts of manually annotated data and trained aesthetic assessment models. To alleviate the dependence on data scale for model training and the biases introduced by trained models, we propose a novel prompt optimization framework, designed to rephrase a simple user prompt into a sophisticated prompt to a text-to-image model. Specifically, we employ the large vision language models (LVLMs) as the solver to rewrite the user prompt, and concurrently, employ LVLMs as a reward model to score the aesthetics and alignment of the images generated by the optimized prompt. Instead of laborious human feedback, we exploit the prior knowledge of the LVLM to provide rewards, i.e., AI feedback. Simultaneously, the solver and the reward model are unified into one model and iterated in reinforcement learning to achieve self-improvement by giving a solution and judging itself. Results on two popular datasets demonstrate that our method outperforms other strong competitors.