Research
Duy Nguyen develops multimodal learning methods for medical and molecular data. His work includes aligning medical images with language and combining molecular fragments with ensembles of three-dimensional conformations for property prediction. Across these applications, he investigates how relationships between different data representations can improve learning and make better use of available training data.
Selected publications
- FACET: A Fragment-Aware Conformer Ensemble Transformer 2026
- ExGra-Med: Extended Context Graph Alignment for Medical Vision-Language Models 2025
Information reviewed 26 September 2026 · University profile