Optimizing Prompts for Text-to-Image Generation

TL;DR

This paper introduces a prompt adaptation framework that automatically refines user prompts for text-to-image models, improving image quality and alignment with user intent through supervised fine-tuning and reinforcement learning.

Contribution

It presents a novel, automated prompt adaptation method combining supervised fine-tuning and reinforcement learning to enhance text-to-image generation.

Findings

Outperforms manual prompt engineering in automatic metrics

Reinforcement learning further improves out-of-domain prompt results

Achieves higher human preference ratings

Abstract

Well-designed prompts can guide text-to-image models to generate amazing images. However, the performant prompts are often model-specific and misaligned with user input. Instead of laborious human engineering, we propose prompt adaptation, a general framework that automatically adapts original user input to model-preferred prompts. Specifically, we first perform supervised fine-tuning with a pretrained language model on a small collection of manually engineered prompts. Then we use reinforcement learning to explore better prompts. We define a reward function that encourages the policy to generate more aesthetically pleasing images while preserving the original user intentions. Experimental results on Stable Diffusion show that our method outperforms manual prompt engineering in terms of both automatic metrics and human preference ratings. Moreover, reinforcement learning further boosts performance, especially on out-of-domain prompts. The pretrained checkpoints are available at https://aka.ms/promptist. The demo can be found at https://aka.ms/promptist-demo.