Language systems for specialized domains

My current focus: NLP and LLM systems that reason over biomedical text, structured clinical data, and domain-specific knowledge to support scientific discovery and clinical decision-making.

• —Biomedical question answering
• —Clinical language understanding
• —Natural-language interfaces to electronic health records
• —Biomedical knowledge grounding
• —Clinical and biomedical LLM systems
• —Multimodal and structured clinical data

Across domains, I study what it takes to turn a general-purpose language model into a system that can be trusted in a specialized setting.

• —Retrieval and knowledge grounding
• —Domain-aware reasoning
• —Evaluation of LLM behavior
• —Explainability and reliability
• —LLM agents and decision-support systems

Before concentrating on biomedical NLP, I studied how NLP systems must be adapted and evaluated under the constraints of other specialized domains — each project shaped how I think about reliability, evaluation, and domain knowledge.

• —Food computing
• —Cross-cultural cuisine transfer and evaluation
• —Phishing and scam detection
• —Automated counter-scam response systems