Recent years have witnessed a rapid growth of deep generative models, with
text-to-image models gaining significant attention from the public. However,
existing models often generate images that do not align well with human
aesthetic preferences, such as awkward combinations of limbs and facial
expressions. To address this issue, we collect a dataset of human choices on
generated images from the Stable Foundation Discord channel. Our experiments
demonstrate that current evaluation metrics for generative models do not
correlate well with human choices. Thus, we train a human preference classifier
with the collected dataset and derive a Human Preference Score (HPS) based on
the classifier. Using the HPS, we propose a simple yet effective method to
adapt Stable Diffusion to better align with human aesthetic preferences. Our
experiments show that the HPS outperforms CLIP in predicting human choices and
has good generalization capability towards images generated from other models.
By tuning Stable Diffusion with the guidance of the HPS, the adapted model is
able to generate images that are more preferred by human users.

为了更好地满足人类审美偏好，本文通过引入人类评分获得的数据集和人类偏好分类器，提出了一种简单有效的基于稳定扩散的方法，通过调整生成模型，生成更符合人类审美偏好的图像。