Author name: Xiangpeng Yang, Ji Xie, Yiyuan Yang, Yue Ma, Yan Huang, Min Xu, Qiang Wu

arXiv:2512.07469v2 Announce Type: replace
Abstract: Existing video editing methods face a critical trade-off: expert models offer precision but rely on task-specific priors like masks, hindering unification; conversely, unified temporal in-context lea…