Context as Memory: Scene-Consistent Interactive Long Video Generation with Memory Retrieval

作者：Jiwen Yu, Jianhong Bai, Yiran Qin, Quande Liu, Xintao Wang, Pengfei Wan, Di Zhang, Xihui Liu · 单位：JIWEN YU∗, The University of Hong Kong, China, JIANHONG BAI∗, Zhejiang University, China, YIRAN QIN, The University of Hong Kong, China, XIHUI LIU†, The University of Hong Kong, China, come world models capable of modeling reality [OpenAI 2024 · 来源：arXiv · 发布日期：2025/10