methods Contextualized Models Models with parameters that adapt to patient, task, or environmental context to support personalized inference. Foundation models as context Using foundation models as flexible priors and contextual memory for downstream inference. Intepretability and Modularity Developing formal understanding of interpretability through structured models, GAMs, and interaction effects. Efficient Foundation Models Structuring context improves inference within foundation models, just as it does in classical statistical models. applications Personalized Genomics Dissecting the biological bases of disease Real-world precision medicine Translating statistical models into useful clinical tools by learning from real-world, heterogeneous patient data. Maternal health Learning from real-world pregnancy data to improve maternal care.
