← Back to Model Beat
Research·1d ago·all news from July 29, 2026

Wonder: Video World Model Done Better

Researchers have introduced Wonder, a video world model that transforms static images or short clips into interactive, camera-controllable environments. By allowing users to navigate and explore these generated spaces in real time, the model improves upon existing techniques for creating immersive synthetic video experiences.

Covered by 1 source

  • AarXiv CS.AIJiacong Xu, Hanwen Jiang, Zhixin Shu, Kalyan Sunkavalli, Vishal M. Patel, Yiqun Mei1d ago

Related stories

ResearchMemory Efficient Audio Synthesis with Decoupled Temporal Depth Diffusion TransformersJul 28 · 2 sourcesResearchAccelerating scientific discovery with ChatGPT for Academic ResearchersJul 29ResearchHow enabling two settings tripled our scores on the ARC-AGI-3 benchmarkJul 29ResearchFrontier AI developers urge international coordination to pace automated research before capabilities outstrip controlJul 29