1 paper
Zhongjie Ba, Shengwang Xu, Peng Cheng +4
Embodied intelligence and world models require video understanding systems to go beyond recognizing objects and actions and develop an understanding of physical regularities. Howev…