titleLooped World Models
authorsHongyuan Adam Lu, Z. L. Victor Wei, Qun Zhang, Jinrui Zeng, Bowen Cao, Lingwei Meng, Mocheng Li, Zezhong Wang, Haonan Yin, Naifu Xue, Minyu Chen, Cenyuan Zhang, Zefan Zhang, Hao Wei, Jiawei Zhou, Haoran Xu, Hao Yang, Ronglai Zuo, Tongda Xu, Yonghao Li, Jian Chen, Hebin Wang, Zeyu Gao, Yang Li, Wei Zhao, Qimin Zhong, Siqi Liu, Yumeng Zhang, Leyan Cui, Zhangyu Wang, Wai Lam
date2026-06-16
venuearXiv (preprint, not peer reviewed)
idarXiv:2606.18208
affiliationFaceMind
upvotes51 (Hugging Face Daily Papers, 2026-11-01)
linksabs pdf hf
primerWorld models
abstractCurrent world models face a fundamental tension: faithful long-horizon simulation demands deep computation, but deeper models are expensive to deploy and prone to compounding errors. We resolve this by introducing Looped World Models (LoopWM), which are the first looped architectures for world modelling. Our method iteratively refines latent environment states through a parameter-shared transformer block. This yield up to 100x parameter efficiency over conventional approaches with adaptive computation that automatically scales depth to match the complexity of each prediction step. Orthogonal to scaling model size and training data, LoopWM establishes iterative latent depth as a new scaling axis for world simulation, which might significantly push the community forward.

cite

@misc{lu2026looped,
  title={Looped World Models},
  author={Hongyuan Adam Lu and Z. L. Victor Wei and Qun Zhang and Jinrui Zeng and Bowen Cao and Lingwei Meng and Mocheng Li and Zezhong Wang and Haonan Yin and Naifu Xue and Minyu Chen and Cenyuan Zhang and Zefan Zhang and Hao Wei and Jiawei Zhou and Haoran Xu and Hao Yang and Ronglai Zuo and Tongda Xu and Yonghao Li and Jian Chen and Hebin Wang and Zeyu Gao and Yang Li and Wei Zhao and Qimin Zhong and Siqi Liu and Yumeng Zhang and Leyan Cui and Zhangyu Wang and Wai Lam},
  year={2026},
  eprint={2606.18208},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2606.18208}
}