1 paper
Chaoyue Li, Yongxue Xu, Jie Feng +1
Recent large multimodal models (LMMs) have become increasingly capable on image and video understanding, yet still struggle to sustain 4D continuous spatiotemporal dynamic reasonin…