1 paper
Yi Yao, Chan-Feng Hsu, Jhe-Hao Lin +5
In spite of recent advancements in text-to-image generation, limitations persist in handling complex and imaginative prompts due to the restricted diversity and complexity of train…