Biography

I am a postdoctoral researcher at UMIACS, working under the supervision of Professor Heng Huang. I received my Ph.D. in Electrical and Computer Engineering from the University of Utah, where I was advised by Professor Yi Zhou. Previously, I completed an M.A. in Statistics at the University of California, Santa Barbara, and earned my B.S. in Statistics at Sichuan University.

For more information, refer to my CV or research statement.

📊 Research Interests

• Memory-efficient algorithms for LLM fine-tuning: I develop and evaluate memory-efficient fine-tuning techniques (e.g., LoRA and Zeroth-Order Optimization) to reduce GPU memory footprint while preserving model performance.

• Robust Optimization & RL Theory: I develop scalable algorithms for reinforcement learning and large-scale optimization, focusing on robustness against uncertainties arising from dependent data and dynamic environments.

• AI for Science: I previously collaborated with researchers at Lawrence Livermore National Laboratory to integrate deep learning with traditional scientific computing software. I am currently extending this line of research at UMD, focusing on addressing non-differentiability in scientific pipelines using modern machine learning techniques.

🔥 News

• 2025.09: One paper selected as a spotlight by NeurIPS 2025!
• 2025.04: Best Reviewer Award by AISTATS 2025!
• 2025.02: One paper selected as a spotlight by ICLR 2025!
• 2024.06: Joined UMD as a postdoctoral researcher!