I am a third-year PhD student in Computer Science at the University of Massachusetts Amherst, co-advised by Dr. Madalina Fiterau in the Information Fusion lab and Dr. Anna Green in the Sequence Analysis and Genomics (SAGE) lab. I am currently a Machine Learning Research Co-op in Synthetic Medicine Design at Biogen, where I collaborate with Dr. Ye Wang.

My research focuses on foundation models, representation learning, generative modeling, model adaptation, and agentic systems. Molecular design and drug discovery are my primary application domains and provide challenging settings with large search and design spaces, expensive evaluation oracles, and limited labeled data. My recent work includes pretrained embedding-based retrieval and molecular generation, reward-guided post-training across autoregressive, masked-diffusion, and discrete-flow generative models, and active-learning workflows for large-scale scientific search. More broadly, I am interested in learning methods that transfer across models and tasks, particularly in settings involving heterogeneous data, expensive feedback, and iterative interaction with external tools.

My current work is highlighted on the Selected Projects page.

Before coming to UMass Amherst, I completed my Master’s degree with distinction in Applied Computational Science and Engineering from Imperial College London in 2024. I completed my master’s thesis, “Hybrid CNN with Multimodal Data for Early Alzheimer’s Disease Forecasting,” with the supervision of Dr. Madalina Fiterau and Dr. James Percival. I obtained a B.E. degree with distinction in Computer Science and Technology at China Agricultural University in 2023, and exchanged to the University of California San Diego in 2021 (University and Professional Studies program).

News

2026.9 Our paper, “Elite-Weighted Supervised Fine-tuning for Goal-Directed Molecular Optimization,” is now available on arXiv.

2026.9 Our review, “Machine Learning-Aided Small-Molecule Virtual Screening: Recent Advances and Future Perspectives,” was published in WIREs Computational Molecular Science.

2026.5 “Advancing Ligand-based Virtual Screening and Molecular Generation with Pretrained Molecular Embedding Distance,” is available on arXiv, also accepted to the ICML 2026 AI4Science workshop.

2026.3 “Mycopermenet-v2: Improved Prediction of Mycomembrane Permeation Using Fusion Noisy Student self-disTillation” was published in the Journal of Chemical Information and Modeling. Also presented this work at ACS Spring 2026, Machine Learning and AI for Organic Chemistry session.

2026.1 — Started internship at Biogen in Cambridge as a Machine Learning Research Intern (Co-op).

2025.01 - Awarded Paul Utgoff Memorial Graduate Scholarship in Machine Learning at UMass Amherst.

2024.11 - Presenting a poster at New England Computer Vision (NECV) Workshop 2024, Yale University, New Haven, CT.