Vibes Meet Gravity: AI Agents Step Into the Real World

Written by Yongyuan Liang and Ruihan Yang
April 06, 2026

Introduction

Code as Interface

Action as Interface

Virtual and Physical: Shared Lessons and Open Gaps

Final Remarks

BibTeX

If you find this work useful, please cite:

@misc{yang2025magmafoundationmodelmultimodal,
      title={Magma: A Foundation Model for Multimodal AI Agents}, 
      author={Jianwei Yang and Reuben Tan and Qianhui Wu and Ruijie Zheng and Baolin Peng and Yongyuan Liang and Yu Gu and Mu Cai and Seonghyeon Ye and Joel Jang and Yuquan Deng and Lars Liden and Jianfeng Gao},
      year={2025},
      eprint={2502.13130},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2502.13130}, 
}

@misc{liang2026anticipatoryplanningmultimodalai,
    title={Anticipatory Planning for Multimodal AI Agents}, 
    author={Yongyuan Liang and Shijie Zhou and Yu Gu and Hao Tan and Gang Wu and Franck Dernoncourt and Jihyung Kil and Ryan A. Rossi and Ruiyi Zhang},
    year={2026},
    eprint={2603.16777},
    archivePrefix={arXiv},
    primaryClass={cs.AI},
    url={https://arxiv.org/abs/2603.16777}, 
}