Any Time, Any Space: A World-State-Centric Framework for Agentic Video Generation

Researchers have developed Any Time, Any Space (ATAS), a framework for agentic video generation that compiles screenplays into shot requests with explicit location, camera, and event-time references. ATAS retrieves evidence from scene and event memories before generation, reconstructs and registers clips after generation, and updates memory only when review and state checks accept observations. Across four video generators, ATAS improves all reported cross-shot consistency measures and achieves the highest aggregate benchmark score for each backbone, outperforming compared agents and 3D-guided methods on target-image spatial revisit and leading three event-recapture criteria.