Any Time, Any Space：以世界状态为中心的智能体视频生成框架

View PDF HTML (experimental)

Abstract:Long-form agentic video generation must turn a screenplay into shots that remain spatially and temporally consistent across nonconsecutive requests. Passing prompts or recent frames between shots does not provide a shared spatial frame for revisiting a scene or a time-indexed record for replaying an earlier event. We present Any Time, Any Space (ATAS), which compiles a screenplay into shot requests with explicit location, camera, and event-time references. Before generation, it retrieves relevant evidence from a scene memory of registered static geometry and an event memory of dynamic observations indexed by scene, event, and time. After generation, ATAS reconstructs and registers the clip. It updates memory only when review and state checks accept the observation, while film delivery remains independent of memory admission. Across four video generators, ATAS improves all reported cross-shot consistency measures and achieves the highest aggregate benchmark score for each backbone. It also outperforms the compared agents and 3D-guided methods on target-image spatial revisit, leads three event-recapture criteria, and receives the highest human ratings on all six criteria. The video results and code are available on our project page: this https URL.

Comments:	Project page: this https URL, Code: this https URL
Subjects:	Computer Vision and Pattern Recognition (cs.CV)
Cite as:	arXiv:2610.20571 [cs.CV]
	(or arXiv:2610.20571v1 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2610.20571 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Weijie Wang [view email]
[v1] Wed, 14 Oct 2026 17:54:18 UTC (20,280 KB)