Learning objectives

• Understand why average accuracy alone is insufficient for deploying medical foundation models in high-stakes biomedical and clinical workflows.
• Learn how conformal selection/alignment can identify high-quality predictions suitable for action while controlling false discoveries.
• Describe the “act-or-defer” framework: FDR-controlled actions for confident cases and calibrated prediction sets for deferred cases.
• Discuss applications to medical AI tasks such as radiology, pathology, biomarker prediction, tumor subtyping, and prognosis.

Speaker Bio

Ying Jin is an Assistant Professor in Statistics and Data Science at the Wharton School, University of Pennsylvania. Prior to that, she was a Wojcicki-Troper Postdoctoral Fellow at Harvard Data Science Initiative from 2024 to 2025. She obtained her PhD in Statistics from Stanford University in 2024, advised by Professors Emmanuel Candès and Dominik Rothenhäusler. Her research centers around uncertainty quantification for black-box AI models, generalizability, distributional robustness, causal inference, and their applications in biomedical discovery and human decisions.