A Definition and Roadmap for World Models
Artificial Intelligence
2026-07-07 v1
Abstract
World models -- internal simulators that learn the structure and dynamics of an environment -- have become one of the most actively debated concepts in AI. From model-based reinforcement learning and video generation to embodied robotics and ultimately, physical AI, researchers across AI subfields are building systems that they call "world models", yet there is no consensus on what a world model fundamentally is, what it should predict, or how it should be built. This perspective article provides a scientific definition of world models, discussions of their key technical aspects, and a staged roadmap for developing effective world models.
Cite
@article{arxiv.2607.06401,
title = {A Definition and Roadmap for World Models},
author = {Xinyuan Chen and Haoyu Guo and Shi Guo and Bingqi Jiang and Chunhua Shen and Xing Shen and Tianfan Xue and Yufei Xue and Mulin Yu and Weinan Zhang and Bin Zhao and Bowen Zhou and Ming Zhou},
journal= {arXiv preprint arXiv:2607.06401},
year = {2026}
}
Comments
Technical report, 58 pages, 10 figures