I started my Ph.D. in Computational Biology and Bioinformatics at Duke University, conducting research under the guidance of Prof. Anru Zhang.
Ph.D. Student · Computational Biology and Bioinformatics · Duke University
I study how models remain reliable under uncertainty.
I am a Ph.D. student in Computational Biology and Bioinformatics at Duke University, conducting research under the guidance of Prof. Anru Zhang. I develop reliable representation learning methods for longitudinal and multimodal data that are irregular, incomplete, and rarely clean.
Research
Health records capture the process of care, not the disease itself.
What gets recorded, and when, reflects how patients move through the health system as much as how they are doing. I develop machine learning methods for longitudinal and multimodal clinical data, including EHR time series, clinical notes, and medical images, drawing on representation learning, causal inference, and statistics.
-
Longitudinal patient representations
Learning representations of patients from irregular, multimodal EHR histories to support downstream clinical prediction.
-
Informative missingness
Treating which measurements, notes, and modalities get recorded as signal rather than noise.
-
Reliable clinical prediction
Calibration, robustness under distribution shift, and causal reasoning so that predictions can inform clinical decisions.
Recent News
Our paper “Two-Sided State-Space Models for Sequential Recommendation with Non-Random Multimodal Review Feedback” has been accepted to Findings of EMNLP 2026.
I graduated from Emory University with a B.S. in Applied Mathematics and Statistics (Highest Honors). Many thanks to my advisor, Prof. Ruoxuan Xiong, for her guidance.
Two papers accepted to Findings of ACL 2026: “Learning Dynamic Representations and Policies from Multimodal Clinical Time-Series with Informative Missingness” and “DART: Mitigating Harm Drift in Difference-Aware LLMs via Distill-Audit-Repair Training.”
Our paper “MambaDATG: Domain-Adaptive Tri-Plane-Gated Pre-training for 3D Abdominal Segmentation” has been accepted as an oral presentation at ICASSP 2026.
Beyond Research
All experiences-
AI literacy
Founded the Emory Artificial Intelligence and Data Association and its “AI for ALL” initiative, growing it to 200+ members across majors.
-
Community
Led a 100+ person team as President of the Emory Chinese Student Association, with programs reaching 2,000+ students and families.
-
Teaching
Served as a teaching assistant for an undergraduate machine learning course at Emory and for Olympiad algebra at AwesomeMath.