Tuchel and Me!

• I obtained my Ph.D. degree in Apex Lab. at Shanghai Jiao Tong University, supervised by Prof. Weinan Zhang and Prof. Yong Yu. I used to be the group leader of ApexRL, the RL research group of Apex Lab. I was also a member of Wu Wen Jun Honorary Doctoral Plan (吴文俊荣誉博士班), advised by Prof. Cewu Lu. From Sept. 2023 - Aug. 2024, I visited Prof. Xiaolong Wang’s Lab at UC San Diego, CA, USA, and did robotics learning research. I won ByteDance Scholars Program (字节跳动奖学金) (10 candidates worldwide each year) in 2022. I received my Bachelor’s Degree at Southwest Jiao Tong University in June 2019. [My Curriculum Vitae] (updated in Dec. 2024)

• My research interests lie in the general area of decision modeling. Particularly, I was devoted to (data-driven) reinforcement learning (RL) methods (such as imitation learning (IL), offline RL) and also foundation models that can help resolve real-world decision-making challenges.

• Recently, I have been working on one of the most complicated decision-making problems, robots, aiming at taking intelligent robots closer to real life, e.g., expanding the skills of robots, and helping robots make smart decisions. In detail, my works focus on how to generate and utilize robot data to control robots, such as sim2real, robot teleoperation, imitation, and real2sim. I believe we should construct a data-efficient robot training and evaluation cycle, and develop the robot control system in a well-designed hierarchical way. I do not believe in fully end2end learning in robot control.

• When I read interesting papers, I sometimes write down notes in Chinese at Zhihu.

Recent News

• Apr. 11, 2025. HugWBC accepted by RSS 2025.
• Sept. 26, 2024. MADiff accepted by NeurIPS 2024.
• Sept. 6, 2024. Three papers (VBC, ACE, GenSim2) accepted by CoRL 2024.
• Jan. 21, 2024. Two papers accepted by ICLR 2024.

Selected Projects (Full Paper List)

* Equal Contribution ^ Project Lead

HugWBC: A Unified and General Humanoid Whole-Body Controller for Fine-Grained Locomotion
Yufei Xue, Wentao Dong, Minghuan Liu^, Weinan Zhang, Jiangmiao Pang
RSS 2025. [paper] [project]

RoboVLMs: Towards Generalist Robot Policies: What Matters in Building Vision-Language-Action Models
Xinghang Li^, Peiyan Li, Minghuan Liu^, Dong Wang, Jirong Liu, Bingyi Kang, Xiao Ma, Tao Kong, Hanbo Zhang^, Huaping Liu
Arxiv 2024. [paper] [project] [code]

ACE: A Cross-platform Visual-Exoskeletons for Low-Cost Dexterous Teleoperation
Shiqi Yang, Minghuan Liu, Yuzhe Qin, Runyu Ding, Jialong Li, Xuxin Cheng, Ruihan Yang, Sha Yi, Xiaolong Wang
CoRL 2024. [paper] [project] [code]

Visual Whole-Body Control for Legged Loco-Manipulation
Minghuan Liu*, Zixuan Chen*, Xuxin Cheng, Yandong Ji, Ruihan Yang, Xiaolong Wang.
CoRL 2024. (Oral). [paper] [project] [code]

GenSim2: Scaling Robot Data Generation with Multi-modal and Reasoning LLMs
Pu Hua*, Minghuan Liu*, Annabella Macaluso*, Yunfeng Lin, Weinan Zhang, Huazhe Xu, Lirui Wang.
CoRL 2024. [paper] [project] [code]

Vision-Language Foundation Models as Effective Robot Imitators
Xinghang Li*, Minghuan Liu*, Hanbo Zhang, Cunjun Yu, Jie Xu, Hongtao Wu, Chilam Cheang, Ya Jing, Weinan Zhang, Huaping Liu, Hang Li, Tao Kong.
ICLR 2024. (Spotlight). [paper] [code] [project]

Visual Imitation with Patch Rewards
Minghuan Liu, Tairan He, Weinan Zhang, Shuicheng Yan, Zhongwen Xu.
ICLR 2023. [paper] [code] [project]

PerfectDou: Dominating DouDizhu with Perfect Information Distillation
Yang Guan*, Minghuan Liu*, Weijun Hong, Weinan Zhang, Fei Fang, Guangjun Zeng, Yue Lin.
NeurIPS 2022. [paper] [code]

Plan Your Target and Learn Your Skills: Transferable State-Only Imitation Learning via Decoupled Policy Optimization.
Minghuan Liu, Zhengbang Zhu, Yuzheng Zhuang, Weinan Zhang, Jianye Hao, Yong Yu, Jun Wang.
ICML 2022. [paper] [code] [project]

Curriculum Offline Imitation Learning
Minghuan Liu*, Hanye Zhao*, Zhengyu Yang, Jian Shen, Weinan Zhang, Li Zhao, Tie-Yan Liu.
NeurIPS 2021. [paper] [code]

Energy-Based Imitation Learning
Minghuan Liu, Tairan He, Minkai Xu, Weinan Zhang.
AAMAS 2021. (Oral). [paper][code]

Competitions

• TOP1, Ubiquant Retro Snake Challenge. [News] May 2022.
• TOP5, Liver Cancer Imaging AI Diagnosis. [Link][News] Apr 2019.
• TOP 6, Finalist of Sports Analytics Challenge (sponsored by PSG), July 2019
• TOP10, SCADA Data Missing Repair Competition. [Link][News] Mar 2019.
• TOP3, Weather Forecast Competition of AI Challenger 2018. [Link][News] Nov 2018.

Honors and Awards

• ByteDance Scholars Program (字节跳动奖学金), Oct. 2022.
• NeurIPS 2022 Top Reviewer, Oct. 2022.
• National Doctoral Scholarship, Oct. 2022.

Services

• Conference Reviewer: NeurIPS’21’22’23, ICML’21’22’23, ICLR’23’24, CVPR’24’25, IROS’24, RAL’25, RSS’25

Misc

• I love playing / watching football (soccer), forward, a fan of Shandong Luneng Taishan, a beginner of football analytics with AI.
  • Click [demo-1] and [demo-2] to see some interesting gadgets I’ve made for football analysis during the Sports Analytics Challenge sponsored by PSG.
  • Have kept a document called Football-Analytics-101 on football analytics motivated by NBAStuffer-Analytics-101, in order to provide ordered, collected and detailed, information about football (soccer) analytics for research works and interesting stuff.
• I also play tennis (NTRP 3.0-3.5) in my spare time.