The “Homogenized” Lens, Study Finds AI Image Generators Default to Just 12 Styles
Stable Diffusion XL was used to produce an image based on a text description. The description was used as input for LLaVA. The description was of ...









