Kevin(Jian) Lu

I am a Master's student in Computer Science at Duke University, where I am fortunate to be advised by Professor Anru Zhang. My research interests span three directions:

• AI for Health — electronic health record (EHR) data quality, where I formulate and build automated methods to detect documentation inconsistencies in clinical records.
• Efficient LLM Inference and Explainability — I investigate where and how information is encoded inside large language models, using these insights to improve both interpretability and inference efficiency.
• LLM Agents — text-to-SQL generation, formulated as an LLM-agent workflow that reasons over databases to translate natural-language questions into SQL.

My long-term goal is to build more efficient and principled AI systems through the theoretical study of models such as large language models, ultimately better supporting real-world applications such as biomedical informatics and healthcare.

Previously, I obtained my B.S. in Computer Science from the University of Rochester, where I worked with Professor Jiebo Luo on computer vision methods for biomedical image analysis, and with Professor Christopher Kanan on evaluating multimodal large language models.