CV
Chuyao Fu's Curriculum Vitae.
Basics
| Name | Chuyao Fu (Troy) |
| fucy2023@berkeley.edu | |
| Website | https://chuyaofu.github.io/ |
| GitHub | ChuyaoFu |
| Google Scholar | Chuyao Fu |
Education
-
Sep. 2023 - Jun. 2027 B.Eng. in Electronic Information Engineering
Southern University of Science and Technology (SUSTech)
- GPA: 3.90/4.00 · Rank: 3/46
- Selected Coursework: Data Structures and Algorithm Analysis, Probability and Statistics, Machine Learning, Robotic Motion and Control, Digital Signal Processing, Signals and Systems, Analog Circuits, Digital Circuits
-
Aug. 2026 - Dec. 2026 Visiting Student
University of California, Berkeley
Research Experience
-
Jun. 2025 - Present Undergraduate Researcher
HMI Lab, School of Computer Science, Peking University
Supervisor: Prof. Shanghang Zhang
- Leading Token-World, a policy-oriented world model that predicts action-conditioned dynamics directly in compact VLM visual-token space, avoiding intermediate RGB reconstruction and reducing simulator step time to 0.359 s.
- Leading VA-CLIP, a reference-free verifier that learns action-video compatibility to evaluate and select embodied world-model rollouts.
- Contributed to LJM, an action-conditioned world model that predicts robot-environment interaction latents before generating future videos.
- First author on both manuscripts; responsible for problem formulation, method design, experiments, and manuscript preparation.
-
Sep. 2024 - Present Undergraduate Researcher
Robotics and Computer Vision Lab, SUSTech
Supervisors: Prof. Hong Zhang and Prof. Max Q.-H. Meng
- Proposed ProDrive, a proactive autonomous-driving framework that jointly trains a planner and a BEV world model through ego-environment co-evolution; first author and lead researcher.
- Co-led AxisField-GS, a feed-forward articulation-recovery framework that predicts dense Gaussian-level motion fields and recovers revolute joint axes from single-state semantic 3DGS assets; co-first author.
Selected Publications and Manuscripts
-
ProDrive: Proactive Planning for Autonomous Driving via Ego-Environment Co-Evolution
Chuyao Fu, Shengzhe Gan, Zhuoli Ouyang, Yuhan Rui, Xiaowei Chi, Sirui Han, Jiankun Wang, Hong Zhang
CVPR 2026 GigaBrain Challenge Workshop
-
Learning Action-Video Compatibility as a Reference-Free Verifier for Embodied World Models
Chuyao Fu, Xiaojie Zhang, Xiaowei Chi, Rongyu Zhang, Yuhan Rui, Jiajun Cao, Yike Guo, Sirui Han, Shanghang Zhang
Under review
-
Token-World: Compact VLM Representations for Policy-Oriented World Modeling
Chuyao Fu, Xiaowei Chi, Yuhan Rui, Yu-Kai Wang, Zezhong Qian, Xiaojie Zhang, Yunfan Lou, Kevin Zhang, Kuangzhi Ge, Chak Wing Mak, Zhi Yang Chen, Athena Zhuoming Zhong, Hongyang Cheng, Haoran Li, Yike Guo, Sirui Han, Shanghang Zhang
Submitted to ICRA 2027
-
AxisField-GS: Feed-Forward Articulation Recovery from Single-State Semantic Gaussian Assets
Shengzhe Gan*, Chuyao Fu*, Yugong Wang, Guangcheng Chen, Hong Zhang. *Equal contribution
Under review
-
Mask World Model: Predicting What Matters for Robust Robot Policy Learning
Yunfan Lou, Xiaowei Chi, Xiaojie Zhang, Zezhong Qian, Chengxuan Li, Rongyu Zhang, Yaoxu Lyu, Guoyu Song, Chuyao Fu, Haoxuan Xu, Pengwei Wang, Shanghang Zhang
ICML 2026
-
EchoArena: Mutual Benchmarking of World Models and VLA Policies
Yu-Kai Wang, Kevin Zhang, Xiaowei Chi, Tianxing Chen, Siqiao Huang, Chuyao Fu, Tiecheng Guo, Peidong Jia, Yan Qin, Kuangzhi Ge, Siyuan Qian, Weishi Mi, Zezhong Qian, Jiajun Li, Qingpo Wuwu, Shanghang Zhang, Jian Tang
CVPR 2026 GigaBrain Challenge Workshop
-
FORCE: Efficient VLA Reinforcement Fine-Tuning via Value-Calibrated Warm-up and Self-Distillation
Shuyi Zhang, Yunfan Lou, Hongyang Cheng, Yichen Guo, Chuyao Fu, Yaoxu Lyu, Xiaojie Zhang, Haoran Li, Pengwei Wang, Zhongyuan Wang, Shanghang Zhang
arXiv preprint, 2026
-
LJM: A Latent Joint-conditional Model for Action-Conditioned Embodied World Model
Kuangzhi Ge, Yu-Kai Wang, Yuhan Rui, Xiaojie Zhang, Hao Chen, Jiaming Liu, Chuyao Fu, Zhi Yang Chen, Tianze Zhou, Zhuoqun Wu, Athend Zhuoming Zhong, Chak-Wing Mak, Kevin Zhang, Xiaowei Chi, Shanghang Zhang, Sirui Han
Under review
Industry Experience
-
Jul. 2026 - Present Beijing
Research Intern
Muka Robotics
- Developing representation learning techniques and data cleaning pipeline for foundational world-action-model pretraining.
- Built a physical world model demo in which a robot observes, predicts, and plays tabletop curling against a human.
-
Mar. 2026 - Apr. 2026 Shenzhen
Research Intern, Future Chemistry Department
XtalPi
- Trained and deployed VLA policies, including LingBot-VLA, π0, and π0.5, for robot manipulation in laboratory-automation settings.
- Integrated perception, policy inference, and control on real robot platforms; conducted iterative evaluation and ablations to diagnose deployment failures.
-
Jun. 2025 - Sep. 2025 Beijing
Research Intern, Embodied Multimodal Large Models Research Center
Beijing Academy of Artificial Intelligence
- Processed large-scale embodied datasets, including AgiBot-World and DROID, for VLA pretraining; implemented action normalization, tokenization, and structured training-data conversion.
- Reproduced representative VLA reinforcement-learning and robot-policy post-training frameworks; contributed to FORCE and the cross-embodiment RoboBrain-X0 project.
Selected Projects
-
Aug. 2026 - Present Awesome Representation for World Models
Open-Source Literature Curation
Project Lead
- Currently lead a collaborative survey of representations for world models, coordinating lab members to curate literature around state and transition representations.
- Designed a Cloudflare-based webpage workflow spanning literature review, taxonomy-based classification, and pull-request review for multi-contributor maintenance.
-
Jun. 2025 - Sep. 2025 RoboBrain-X0: Cross-Embodiment Vision-Language-Action Model
Open-Source VLA Model
Contributor
- Contributed to RoboBrain-X0, a unified VLA model for cross-embodiment action generation that supports heterogeneous robot platforms through a shared action-token representation.
- Focused on the data pipeline, processing AgiBot-World and DROID with action normalization, tokenization, and structured conversion for large-scale VLA pretraining.
Leadership and Service
-
2024 - 2025 Class Leader, Class 2303
Class Committee
- Led class operations and activities, including event planning, logistics coordination, routine administration, and award application materials; the class received honors including Excellent Undergraduate Academic Style, Best Sports Team, and Best Popularity Class.
-
2024 - 2025 House Division Lead, Street Dance Club
Student Organization
- Led the House division by coordinating weekly training with instructors and organizing two campus street dance performances, including rehearsal planning and member coordination.
Honors and Awards
-
Sep. 2025 First Prize, Outstanding Student Award
SUSTech (top 5%; RMB 5,000 scholarship)
-
Sep. 2024 Third Prize, Outstanding Student Award
SUSTech (top 25%; RMB 1,000 scholarship)
-
Sep. 2023 First-Class Admission Scholarship
SUSTech (top 10%; RMB 50,000 scholarship)
-
Nov. 2022 Gold Medal, Chinese Chemistry Olympiad
Chinese Chemical Society (top 150 nationwide)