Lai Wei

| Email | Blog Posts|
| Github | Twitter | LinkedIn |

Hi! I am Lai Wei, a second-year master student in Computer Science & Engineering at UC San Diego, fortunately advised by Prof. Xiaolong Wang.

Previously, I received my Bachelor's degree in CSE from CUHK-Shenzhen, where I was advised by Prof. Ruimao Zhang, Prof. Tin Lun Lam, and Prof. Xiaoguang Han.

My research interests lie in the intersection of Robotics and Computer Vision. I am particularly interested in humanoid loco-manipulation and dexterous grasping/tool use, and on merging these capabilities to enable force-aware robotic manipulation in contact-rich environments.

I'm actively applying for Ph.D. positions for Fall 2026.


  Publications

From Power to Precision: Learning Fine-grained Dexterity for Multi-fingered Robotic Hands
Jianglong Ye*, Lai Wei*, Guangqi Jiang, Changwei Jing, Xueyan Zou, and Xiaolong Wang
arXiv preprint, 2025

webpage | arXiv | hardware | video | paper | twitter | bibtex
          
@misc{ye2025powerprecisionlearningfinegrained,
      title={From Power to Precision: Learning Fine-grained Dexterity for Multi-fingered Robotic Hands}, 
      author={Jianglong Ye and Lai Wei and Guangqi Jiang and Changwei Jing and Xueyan Zou and Xiaolong Wang},
      year={2025},
      eprint={2511.13710},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2511.13710}, 
}
        

HMC: Learning Heterogeneous Meta-Control for Contact-Rich Loco-Manipulation
Lai Wei*, Xuanpin Peng*, Ri-Zhao Qiu, Tianshu Huang, Xuxin Cheng, and Xiaolong Wang
RSS 2025 Workshop on WCBM Spotlight

webpage | arXiv | paper | twitter | bibtex
          
@misc{wei2025hmclearningheterogeneousmetacontrol,
      title={HMC: Learning Heterogeneous Meta-Control for Contact-Rich Loco-Manipulation}, 
      author={Lai Wei and Xuanbin Peng and Ri-Zhao Qiu and Tianshu Huang and Xuxin Cheng and Xiaolong Wang},
      year={2025},
      eprint={2511.14756},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2511.14756}, 
}
        

In-N-On: Scaling Egocentric Manipulation with In-the-Wild and On-Task Data
Xiongyi Cai*, Ri-Zhao Qiu*, Geng Chen, Lai Wei, Isabella Liu, Tianshu Huang, Xuxin Cheng, Xiaolong Wang
arXiv preprint, 2025

webpage | arXiv | code | model | data | bibtex
@misc{cai2025innon,
      title={In-N-On: Scaling Egocentric Manipulation with In-the-Wild and On-Task Data},
      author={Xiongyi Cai and Ri-Zhao Qiu and Geng Chen and Lai Wei and Isabella Liu and 
              Tianshu Huang and Xuxin Cheng and Xiaolong Wang},
      year={2025},
      eprint={2511.15704},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2511.15704},
}
      

Ensuring Force Safety in Vision-Guided Robotic Manipulation via Implicit Tactile Calibration
Lai Wei*, Jiahua Ma*, Yibo Hu, and Ruimao Zhang
Conference on Robot Learning (CoRL), 2025

webpage | arXiv | paper | bibtex
          
@misc{wei2024ensuringforcesafetyvisionguided,
      title={Ensuring Force Safety in Vision-Guided Robotic Manipulation via Implicit Tactile Calibration}, 
      author={Lai Wei and Jiahua Ma and Yibo Hu and Ruimao Zhang},
      year={2024},
      eprint={2412.10349},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2412.10349}, 
}
        

Online Dual Robot-Human Collaboration Trajectory Generation by Convex Optimization
Lai Wei, Yanzhe Wang, Yibo Hu, Tin Lun Lam, and Yanding Wei
Robotics and Computer-Integrated Manufacturing, 2025

Journal link | paper | video | bibtex
          
@article{wei2025online,
  title={Online dual robot--human collaboration trajectory generation by convex optimization},
  author={Wei, Lai and Wang, Yanzhe and Hu, Yibo and Lam, Tin Lun and Wei, Yanding},
  journal={Robotics and Computer-Integrated Manufacturing},
  volume={91},
  pages={102850},
  year={2025},
  publisher={Elsevier}
}
        

SCoDA: Domain Adaptive Shape Completion for Real Scans
Yushuang Wu, Zizheng Yan*, Ce Chen, Lai Wei, Xiao Li, Guanbin Li, Yihao Li, Shuguang Cui, and Xiaoguang Han
Conference on Computer Vision and Pattern Recognition (CVPR), 2023

webpage | arXiv | code | paper | bibtex
          
@inproceedings{wu2023scoda,
  title={Scoda: Domain adaptive shape completion for real scans},
  author={Wu, Yushuang and Yan, Zizheng and Chen, Ce and Wei, Lai and Li, Xiao and Li, Guanbin and Li, Yihao and Cui, Shuguang and Han, Xiaoguang},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={17630--17641},
  year={2023}
}
        

An Online Collision-free Trajectory Generation Algorithm for Human–robot Collaboration
Yanzhe Wang, Lai Wei, Kunpeng Du, Gongping Liu, Qian Yang, Yanding Wei, and Qiang Fang
Robotics and Computer-Integrated Manufacturing, 2023

Journal link | paper | video | bibtex
          
@article{wang2023online,
  title={An online collision-free trajectory generation algorithm for human--robot collaboration},
  author={Wang, Yanzhe and Wei, Lai and Du, Kunpeng and Liu, Gongping and Yang, Qian and Wei, Yanding and Fang, Qiang},
  journal={Robotics and Computer-Integrated Manufacturing},
  volume={80},
  pages={102475},
  year={2023},
  publisher={Elsevier}
}
        

  Academic Services

Teaching Assistant

  • Signals and Systems (CUHK-Shenzhen ECE3001, SP 24)
  • Data Structures (CUHK-Shenzhen CSC3100, FA 23)
  • Probability and Statistics (CUHK-Shenzhen STA2001, SP 23)
  • C/C++ Programming (CUHK-Shenzhen CSC3002, FA 22)

Conference Reviewer

  • Computer Vision and Pattern Recognition (CVPR, 2026)


Website template from here