9 citations · 31 across the 34 of their papers we have counts for
1 paper · 1 filter
Tao Feng, Xianbing Zhao, Zhenhua Chen +4
Recent advances in diffusion-based and autoregressive video generation models have achieved remarkable visual realism. However, these models typically lack accurate physical alignm…