Infinite-World: Scaling Interactive World Models to 1000-Frame Horizons via Pose-Free Hierarchical Memory

Ruiqi Wu, Xuanhua He, Meng Cheng, Tianyu Yang, Yong Zhang, Zhuoliang Kang, Xunliang Cai, Xiaoming Wei, Chunle Guo, Chongyi Li, Ming-Ming Cheng · 2026

Read Infinite-World: Scaling Interactive World Models to 1000-Frame Horizons via Pose-Free Hierarchical Memory by Ruiqi Wu, Xuanhua He, Meng Cheng, Tianyu Yang, Yong Zhang, Zhuoliang Kang, Xunliang Cai, Xiaoming Wei, Chunle Guo, Chongyi Li, Ming-Ming Cheng (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-02393.txt.

Description

Ruiqi Wu and 10 co-authors build an interactive world model that stays coherent for more than a thousand frames, in a 2026 arXiv paper. A world model generates video that responds to a user's actions, like a playable simulated place. Real-world training video is messy: camera poses are noisy and viewpoints rarely repeat, so long-term memory is hard to learn.

Their system, Infinite-World, compresses past frames into a fixed-size memory without needing pose estimates. A second module turns continuous motion into a three-state action label, shielding training from noisy trajectories. A short fine-tuning set, only thirty minutes of video chosen to revisit places often, activates long-range loop closure. Tests and user studies report better visual quality, action control, and spatial consistency than rivals.

Catalog record

CREATORRuiqi Wu, Xuanhua He, Meng Cheng, Tianyu Yang, Yong Zhang, Zhuoliang Kang, Xunliang Cai, Xiaoming Wei, Chunle Guo, Chongyi Li, Ming-Ming Cheng
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE38 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-02393

Browse the full library · About the Tower of Babel