Improving GFlowNets for Text-to-Image Diffusion Alignment

TL;DR

This paper introduces DAG, a GFlowNet-based method to improve text-to-image diffusion models by aligning generated images with desired properties specified through black-box rewards, enhancing control and quality.

Contribution

It proposes a novel post-training approach using GFlowNets to align diffusion models with reward functions without direct reward maximization.

Findings

Effective alignment with black-box rewards demonstrated on Stable Diffusion

Improved image quality and property control over baseline methods

Robust performance across various reward specifications

Abstract

Diffusion models have become the de-facto approach for generating visual data, which are trained to match the distribution of the training dataset. In addition, we also want to control generation to fulfill desired properties such as alignment to a text description, which can be specified with a black-box reward function. Prior works fine-tune pretrained diffusion models to achieve this goal through reinforcement learning-based algorithms. Nonetheless, they suffer from issues including slow credit assignment as well as low quality in their generated samples. In this work, we explore techniques that do not directly maximize the reward but rather generate high-reward images with relatively high probability -- a natural scenario for the framework of generative flow networks (GFlowNets). To this end, we propose the Diffusion Alignment with GFlowNet (DAG) algorithm to post-train diffusion models with black-box property functions. Extensive experiments on Stable Diffusion and various reward specifications corroborate that our method could effectively align large-scale text-to-image diffusion models with given reward information.