Personal Website Google Scholar

I am mentored by Dr. Fabien Scalzo at Pepperdine University, where I work on various AI problems in medicine, including glaucoma diagnostics and 2D to 3D neurovascular reconstruction. More recently, I have become interested in open-source LLMs and how prompt perturbations and adversarial attacks can affect performance. Additionally, I am learning about self-optimization with textual gradient descent at Stanford's AI Lab under Dr. James Zou.

Selected Interesting Papers

• Wu, Sean, Koo, M., Kao, L.Y., Black, A., Blum, L., Scalzo, F., & Kurtz, I. (2024). Adversarial Databases Improve Success in Retrieval-Based Large Language Models. NeurIPS 2024 Workshop in AdvML-Frontiers, arXiv:2407.14609.
• Wu, Sean, Koo, M., Blum, L., Black, A., Kao, L., Fei, Z., Scalzo, F., & Kurtz, I. (2024). Benchmarking Open-Source Large Language Models, GPT-4, and Claude 2 on Multiple-Choice Questions in Nephrology. NEJM AI, 1(2), AIdbp2300092.
• Wu, Sean, Kaneko, N., Liebeskind, D.S., et al. (2024). Fourier Feature Network for 3D Vessel Reconstruction from Biplane Angiograms. Machine Vision and Applications, 35, 106. https://doi.org/10.1007/s00138-024-01585-5

Recent Presentation (2023)

Topic: Auxiliary Domain Learning @ MICCAI Omia-X, Vancouver, Canada

Selected Media Coverage

Press Outlet	Article Title
ZDNET - Artificial Intelligence	Why Open-Source Generative AI Models Are Still a Step Behind GPT-4
Medium	AI in Nephrology: A Comparative Analysis of Leading Large Language Models’ Efficacy in Medical Education and Patient Care
Newswise	Proprietary Large Language Models Outperform Open-Source Counterparts on Multi-Choice Nephrology Questions
Goldwater Foundation	Awarded the Barry Goldwater Scholarship
Computing Research Association	Recognition from the Computing Research Association
Pepperdine University News	Applications of Large Language Models in Healthcare
Pepperdine IT Annual Review	Overview of My Undergraduate Research