1 paper
Zhongang Cai, Ruisi Wang, Chenyang Gu +26
Despite remarkable progress, multimodal foundation models still exhibit surprising deficiencies in spatial intelligence. In this work, we explore scaling up multimodal foundation m…