Introduces a 'Hybrid Memory' architecture that maintains the identity and motion of dynamic subjects even when they hide out of view.
March 27, 2026
Original Paper
Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models
arXiv · 2603.25716
The Takeaway
Current world models often lose track of subjects that temporarily leave the frame; this model acts as both an archivist for static scenes and a vigilant tracker for dynamic subjects. This is crucial for realistic physical simulation and long-term consistency in robotic world models.
From the abstract
Video world models have shown immense potential in simulating the physical world, yet existing memory mechanisms primarily treat environments as static canvases. When dynamic subjects hide out of sight and later re-emerge, current methods often struggle, leading to frozen, distorted, or vanishing subjects. To address this, we introduce Hybrid Memory, a novel paradigm requiring models to simultaneously act as precise archivists for static backgrounds and vigilant trackers for dynamic subjects, en