Author name: Yaohou Fan, Qingzhong Wang, Yongsong Huang, Junyi Liu, Tomo Miyazaki, Shinichiro Omachi

arXiv:2604.24171v1 Announce Type: new
Abstract: Current visual text generation models struggle with the trade-off between text accuracy and overall image coherence. We find that achieving high text accuracy can reduce aesthetic quality and instruction…