COSA: Concatenated Sample Pretrained Vision-Language Foundation Model.

Sihan Chen, Xingjian He, Handong Li, Xiaojie Jin, Jiashi Feng, Jing Liu

Browse the full ICLR paper archive.