-
Geographical Representativeness of Images from Text-to-Image Models
The dataset used in the paper is a crowdsourced study comprising 540 participants across 27 countries. -
RECAP: Principled Recaptioning Improves Image Generation
A text-to-image diffusion model trained on a recaptioned dataset to improve image generation quality and semantic alignment. -
Class-Conditional Self-Rewarding for Text-to-Image Models
Self-rewarding mechanism for Text-to-Image models, using image captioning methods. -
Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Pr...
Text-to-image diffusion models, e.g. Stable Diffusion (SD), lately have shown remarkable ability in high-quality content generation, and become one of the representatives for... -
LAION-Aesthetics2
A dataset of images generated by various text-to-image models, including Stable Diffusion and Midjourney.
