I am a PhD candidate in Computational Mathematics at Fudan University, supervised by Prof. Jin Cheng. For the 2024-2025 academic year, I am visiting the Fields Institute for Research in Mathematical Sciences, working as a research assistant under the supervision of Prof. Huaxiong Huang. My research focuses on Large Language Models, and High Performance Computing. I have experiences in mathematical modeling, scientific computing, and machine learning applications in data-driven science.

🔭 Research Area

• Large Language Models: Reinforcement Learning, Training Acceleration
• AI for Science: Equation Discovery, Dynamic Identification
• High Performance Computing: Data Compression

🎓 Educations

• 2021.09 - 2026.06, Ph.D. in Computational Mathematics, Fudan University, Shanghai, China.
• 2024.05 - 2025.05, Joint Doctoral Programme, The Fields Institute for Research in Mathematical Sciences, Toronto, Canada.
• 2017.09 - 2021.06, B.Sc. in Financial Mathematics, Shanghai University of Finance and Economics, Shanghai, China.

📝 Publications

• Xinyue Luo, Xuzhe Qian, Yu Chen, Huaxiong Huang, Jin Cheng. Reconstructing Dynamics from Steady Spatial Patterns with Partial Observations. International Conference on Learning Representations(ICLR) Workshop XAI4Science, 2025.

• Xinyue Luo, Jin Cheng, Yu Chen. MeLLoC: Lossless Compression with High-order Mechanism Learning. Conference on Neural Information Processing Systems(NeurIPS), 2024.

• Xinyue Luo, Yu Chen, Jin Cheng, Zhaoyi Sun. A Mechanism-Based Method for Image Inpainting. Applied Mathematics Letters, 2024.

• Xinyue Luo, Jin Cheng, Zhaoyi Sun, Yu Chen. Precision-preserving Compression of Scientific Data: Learn Mechanism from Data. 2024 Data Compression Conference, IEEE, 2024.

• Xinyue Luo, Yu Chen. Parameter Identification of Vegetation Pattern Dynamic System. Journal of Fudan University (Natural Science, in Chinese), 2022.

• Xinyue Luo, Nian Shao, Jin Cheng, Wenbin Chen. Modeling the Trend of Outbreak of COVID-19 in the Diamond Princess Cruise Ship Based on a Time-delay Dynamic System. Mathematical Modeling and Its Applications, 2020.

• Yue Yan, Yu Chen, Keji Liu, Xinyue Luo, Boxi Xu, Yu Jiang, Jin Cheng. Modeling and Prediction for the Trend of Outbreak of NCP Based on A Time-delay Dynamic System. Scientia Sinica Mathematica, 2020.

• Hanshuang Pan, Niao Shao, Yue Yan, Xinyue Luo, Shufen Wang, Lin Ye, Jin Cheng, Wenbin Chen. Multi‐Chain Fudan‐CCDC Model for COVID‐19: A Revisit to Singapore’s Case. Quantitative Biology, 2020.

💬 Conferences

• Rethink Compression: Learn Mechanism from Data, June 2024
  • International Conference on Inverse Problems Theory & Computation – Poster Session
• A Mechanism-based Method for Image Inpainting, Apr 2024
  • Joint Fudan - RICAM Seminar on Inverse Problems – Oral Presentation
• Parameter Identification of Reaction-Diffusion Systems, Oct 2023
  • Youth Forum on Mathematics and Data Science – Oral Presentation
• Inverse Problems of Vegetation Pattern Systems, Aug 2023
  • The International Congress on Industrial and Applied Mathematics(ICIAM) 2023 – Poster Session
• Advancing Data Compression via Mechanism Learning, Mar 2023
  • Complex Systems Science and Applications – Poster Session

🎖 Honors and Awards

• Scholar Award, Conference on Neural Information Processing Systems (NeurIPS), 2024
• Best Poster Award, Complex Systems Science and Applications, 2023
• First Place, National Industrial and Financial Modeling Competition, 2021
• First Prize, National MathorCup Mathematical Modeling Challenge, 2020
• Second Prize, National Mathematical Modeling Competition, 2018 & 2019

💻 Internships

• 2025.06 - now, LLM research Intern (Reinforcement Learning), Ant Group(Plan A - Talent Programme).
• 2022.03 - 2024.03, MATLAB Ambassador, MathWorks.
• 2021.03 - 2021.06, NLP Research Intern, Ping An Technology.

🏅 Scholarships

• National Scholarship (Top 0.2%), 2024
• Lu Zonglin Scholarship (Top 5 University-wide), 2023
• Fudan University Willingdon & Partners Named Scholarship, 2022

Contact

If you are interested in my research or potential collaborations, feel free to reach out!