Zhongwei Wan

Zhongwei Wan works on large language models and multimodal AI systems, focusing on making these models more efficient (e.g., compressing model size and memory usage), evaluating their reliability (e.g., detecting hallucinated or incorrect outputs, especially in medical settings), and building benchmarks to test reasoning, mathematical, and multimodal capabilities. Much of the work involves applying these methods to domains like medicine (medical question answering, radiology reports, ECG signals) and long-context or agent-based reasoning tasks.

Publication output grew sharply from a handful of papers per year before 2022 to over 20 per year in 2024-2025, indicating a period of rapid and increasing research activity in recent years.

Generated by claude-sonnet-5 from public bibliographic data · Jul 20, 2026