1 paper · 1 filter
Jiaxin Wu, Yihao Pi, Yinling Zhang +2
Generative video models are increasingly studied as implicit world models, yet evaluating whether they produce physically plausible 3D structure and motion remains challenging. Mos…