AI RESEARCH
WISTERIA: Learning Clinical Representations from Noisy Supervision via Multi-View Consistency in Electronic Health Records
arXiv CS.AI
•
ArXi:2605.09765v1 Announce Type: cross Representation learning in electronic health records (EHR) has largely followed paradigms inherited from natural language processing, relying on sequence modeling and reconstruction based objectives that treat clinical labels as ground truth. However, real world clinical supervision is inherently weak, arising from heterogeneous, noisy, and institution specific labeling processes such as billing codes, heuristic phenotypes, and incomplete annotations.