🧑‍🎓 About Me

I am a Ph.D. candidate in Computer and Information Engineering at The Chinese University of Hong Kong, Shenzhen (CUHK-SZ), under the supervision of Prof. Jianfeng Mao and Prof. Wenye Li. I obtained the Bachelor’s degree in Physics from the University of Chinese Academy of Sciences (UCAS) in 2020.

I have a broad interest in machine learning with a particular interest in AI for Science, (Multimodal) Large Language Model, and Statistics for AI. My CV can be found here in Chinese and English (last updated: 2025-05-04).

🏷️ Research Interests

AI for Science, (Multimodal) Large Language Model, Statistics for AI.

📑 Research Topics

• Efficient Similarity and Distance Learning (Previous Work)
  • Optimize similarity matrices for offline and online incomplete data (Published in NeurIPS, WWW, and UAI)
  • Estimate distance matrices for incomplete data in similarity search tasks (Published in AAAI, ECAI, and ECML)
  • Develop an accurate tree-Wasserstein distance for approximating the 1-Wasserstein distance (Published in ICML)
• Physical Reasoning for (Multimodal) Large Language Models (Ongoing Work)
  • Develop a new benchmark to evaluate the physical reasoning abilities of (M)LLMs
  • Design a novel framework to enhance the physical reasoning abilities of (M)LLMs

📝 Publications & Preprint

$\dagger$: Equal contributions; $\ast$: Corresponding authors

• UltraTWD: Optimizing Ultrametric Trees for Tree-Wasserstein Distance
  Fangchen Yu, Yanzhen Chen, Jiaxing Wei, Jianfeng Mao, Wenye Li^$\ast$, Qiang Sun^$\ast$
  [Paper] [Code] [Poster] [bibtex]

• A Theory-Driven Approach to Inner Product Matrix Estimation for Incomplete Data: An Eigenvalue Perspective
  Fangchen Yu^{$\dagger \ast$}, Yicheng Zeng^{$\dagger \ast$}, Jianfeng Mao, Wenye Li
  [Paper] [Code] [Poster] [bibtex]

• Balancing the Trade-Off Between Global and Personalized Performance in Federated Learning
  Zibin Pan, Chi Li, Fangchen Yu, Shuyi Wang, Xiaoying Tang^$\ast$, Junhua Zhao^$\ast$
  [Paper] [bibtex]

• KAE: Kolmogorov-Arnold Auto-Encoder for Representation Learning
  Fangchen Yu, Ruilizhen Hu, Yidong Lin, Yuqi Ma, Zhenghao Huang, Wenye Li^$\ast$
  [Paper] [Code] [bibtex]

• FedLF: Layer-wise Fair Federated Learning
  Zibin Pan, Chi Li, Fangchen Yu, Shuyi Wang, Haijin Wang, Xiaoying Tang^$\ast$, Junhua Zhao^$\ast$
  [Paper] [Code] [Poster] [bibtex]
  

• DocReal: Robust Document Dewarping of Real-Life Images via Attention-Enhanced Control Point Prediction
  Fangchen Yu (internship), Yina Xie, Lei Wu, Yafei Wen, Guozhi Wang, Shuai Ren, Xiaoxin Chen, Jianfeng Mao, Wenye Li^$\ast$
  [Paper] [Code] [Poster] [bibtex]
  

• Boosting Spectral Clustering on Incomplete Data via Kernel Correction and Affinity Learning
  Fangchen Yu, Runze Zhao, Zhan Shi, Yiwen Lu, Jicong Fan, Yicheng Zeng, Jianfeng Mao, Wenye Li^$\ast$
  [Paper] [Code] [Video] [PPT] [Poster] [bibtex]
  

• Online Estimation of Similarity Matrices with Incomplete Data
  Fangchen Yu, Yicheng Zeng, Jianfeng Mao, Wenye Li^$\ast$
  [Paper] [Code] [Poster] [bibtex]
  

• Highly-Efficient Robinson-Foulds Distance Estimation with Matrix Correction
  Fangchen Yu, Rui Bao, Jianfeng Mao, Wenye Li^$\ast$
  [Paper] [Code] [Poster] [bibtex]
  

• From Incompleteness to Unity: A Framework for Multi-view Clustering with Missing Values
  Fangchen Yu, Zhan Shi, Yuqi Ma, Jianfeng Mao, Wenye Li^$\ast$
  [Paper] [bibtex]

• Metric Nearness Made Practical
  Wenye Li, Fangchen Yu, Zichen Ma
  [Paper] [Code] [Poster] [bibtex]
  

• Calibrating Distance Metrics Under Uncertainty
  Wenye Li^$\ast$, Fangchen Yu
  [Paper] [bibtex]

• Learning Sparse Binary Code for Maximum Inner Product Search
  Changyi Ma, Fangchen Yu, Yueyao Yu, Wenye Li^$\ast$
  [Paper] [bibtex] (Best Short Paper Finalist)

🏫 Educations

• Sep. 2020 - Oct. 2025,
  Ph.D. Candidate in Computer and Information Engineering, at The Chinese University of Hong Kong, Shenzhen (CUHK-SZ; Supervisors: Prof. Jianfeng Mao and Prof. Wenye Li).
  

• Sep. 2016 - Jun. 2020,
  B.S. in Physics, at University of Chinese Academy of Sciences (UCAS).
  

• Oct. 2024 - Apr. 2025,
  Visiting Student, at Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI, ranked Top10 globally in AI; Supervisor: Prof. Qiang Sun).
  
• Aug. 2019 - Dec. 2019,
  Global Study Program, at University of California, Davis (UCD).

💻 Internships

• Jun. 2025 - Sep. 2025,
  Research Intern, at Shanghai AI Lab, Shanghai, China. (Supervisor: Dr. Peng Ye)
  
• Sep. 2020 - Aug. 2024,
  Graduate Research Assistant, at Shenzhen Research Institute of Big Data, Shenzhen, China. (Supervisors: Prof. Wenye Li and Prof. Yicheng Zeng)
  
• Mar. 2023 - Jul. 2023,
  Research Intern, at Vivo AI Lab, Shenzhen, China. (Group: Visual Understanding and Generation)

🎖 Honors and Awards

• 2023, PhD Fellowship, Shenzhen Research Institute of Big Data
• 2022, PhD Fellowship, Shenzhen Research Institute of Big Data
• 2022, Outstanding Teaching Assistant Award, The Chinese University of Hong Kong, Shenzhen
• 2021, Outstanding Teaching Assistent Award, The Chinese University of Hong Kong, Shenzhen
• 2021, PhD Fellowship, Shenzhen Research Institute of Big Data
• 2020, PhD Fellowship, Shenzhen Research Institute of Big Data
• 2019, Class III Scholarship, University of Chinese Academy of Sciences
• 2017, Merit Student, University of Chinese Academy of Sciences

👨🏻‍💻 Academic Service

• Program Committee Member:
  ICML 2025, ICLR 2025, NeurIPS 2025/2024, WWW 2025, AAAI 2025/2024/2023, IJCAI 2025/2024, ECML 2022, etc.

👨🏻‍🏫 Teaching Assistant

• MAT3007 Optimization. Instructor: Prof. Junfeng Wu. Summer 2024.
• DDA4210 Advanced Machine Learning. Instructors: Prof. Jicong Fan & Prof. Tongxin Li. Spring & Fall 2023.
• MAT3300 Mathematical Modeling. Instructor: Prof. Gongqiu Zhang. Fall 2022.
• EIE4006 Performance Evaluation of Communication Networks. Instructor: Prof. Tony Lee. Summer 2022.
• STA3010 Regression Analysis. Instructor: Prof. Feng Yin. Spring 2022.
• MAT4003 Number Theory. Instructor: Prof. Jun Bo (Mario) Huan. Fall 2021.
• MAT4004 Graph Theory. Instructor: Prof. Jun Bo (Mario) Huang. Spring 2021.
• MAT3280 Probability Theory. Instructor: Prof. Kenneth Shum. Fall 2020.

Last Update: 2025-08-27