GROME is introduced, a novel factorized memory framework that augments video world models for long-horizon coherence across both static scenes and dynamic entities and readily accommodates external observations as plug-and-play memory, unlocking flexible and controllable video continuation and scene exploration.
Fang-Jie Yu, Xiao-Xue Wu, Hai-Yu Zhang et al.· 0 citations
Autoregressive video generation relies on memory retrieval to model long-range dependencies. This capability becomes particularly critical for multi-shot generation, where models must reconcile substantial visual changes across shot transitions with persistent semantic continuity. Existing methods tend to perform memor...
Xiao-Xue Wu, Xin-Yu Gao, Fang-Jie Yu et al.· 0 citations
Every valid trajectory can teach what physically happened, while the actor is deployed only under the condition associated with relatively better actions, in \method, a direct world-action policy that combines outcome-agnostic predictive learning with outcome-aware policy improvement.
Bo-Tong Zhao, Fangjie Yu, Tim Yu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.