1 paper
Etai Sella, Yanir Kleiman, Hadar Averbuch-Elor
Despite rapid advancements in the capabilities of generative models, pretrained text-to-image models still struggle in capturing the semantics conveyed by complex prompts that comp…