2607.06401

Total: 1

#1 A Definition and Roadmap for World Models [PDF] [Copy] [Kimi] [REL]

Authors: Xinyuan Chen, Haoyu Guo, Shi Guo, Bingqi Jiang, Chunhua Shen, Xing Shen, Tianfan Xue, Yufei Xue, Mulin Yu, Weinan Zhang, Bin Zhao, Bowen Zhou, Ming Zhou

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.

Subject: Artificial Intelligence

Publish: 2026-07-07 15:31:32 UTC