I am a researcher at Microsoft Research, collaborating with Hoifung Poon. My research focuses on multimodal machine learning for clinical diagnosis and precision health.

Research

My research focuses on AI for healthcare. I am interested in applying machine learning to various modalities, such as brain MRI and chest X-ray, to predict radiologic, pathologic, and oncologic findings. My work focuses on real-world evidence derived from electronic health records, which is often heterogeneous and unstructured. I am interested in curating such data at scale and using it to define and solve clinically meaningful research problems.

Meanwhile, I do data analysis for biology problems, collaborating with researchers at Johns Hopkins Medicine and Stanford University and publishing papers in biology journals.

In the past, I worked on efficient language modeling (thesis topic), semantic parsing, prompting (the best short paper in NAACL 2021), time-series analysis, and grounded language acquisition.

A full list of papers can be found in publications.

Engineering

I enjoy implementing and optimizing machine learning algorithms and working with large-scale data. I contributed to PyTorch Lightning and Radiology Report Generation Metrics.

I am also known as hiaoxui, the founder and major maintainer of WallessPKU, a non-profit anti-censorship project that helps Chinese students and researchers to circumvent the Great Firewall. Since 2017, it has been serving more than 64,000 scholars from 200+ institutions.

Education & Experience

I received my Ph.D. in Computer Science in 2024 from Johns Hopkins University, advised by Benjamin Van Durme. My thesis was on natural language processing. Prior to that, I obtained my B.S. in Physics and Computer Science from Peking University in 2019.

I joined Microsoft Research as a researcher in 2024. I previously interned at Microsoft several times (2017 in Beijing, 2022 and 2023 in Redmond). I was a visiting student at Johns Hopkins University in 2018, mentored by Jason Eisner.

Honors

• Outstanding Reviewer for EMNLP 2019, NeurIPS 2025, and ICML 2026.
• Best Short Paper Award at NAACL 2021. [the paper↗]
• Silver Medalist in Chinese Physics Olympiad (CPhO) in 2014.

Teaching & Mentoring

I was a TA for Machine Learning (CS 601.475) at Johns Hopkins University in 2022. I have also mentored:

• Jiachen Tu, research intern at Microsoft Research, 2025. Now Ph.D. student at UIUC.
• Vivek Chari, undergraduate student at Johns Hopkins University, 2023. Now Scientist at Microsoft.
• Yukun Feng, master's student at Johns Hopkins University, 2022. Now Scientist at Apple.