AI RESEARCH
Efficient Generative Prediction for EHR Foundation Models: The SCOPE and REACH Estimators
arXiv CS.LG
•
ArXi:2602.03730v2 Announce Type: replace-cross Generative foundation models trained on tokenized electronic health record (EHR) timelines show promise for clinical outcome prediction via Monte Carlo sampling of simulated future trajectories. However, this approach suffers from three coupled limitations: sparse estimate distributions that poorly differentiate patient risk levels, extreme computational cost, and high sampling variance.