Clinical Cue Calibration — an evaluation apparatus
Abstract
An evaluation apparatus for whether clinical LLMs weight diagnostic evidence correctly. A schema, a paired-probe harness (resist / detect / inquire failure modes), an authoring discipline, and an advisory automated judge. Ships the template plus three demo cases; evaluation sets are clinician-authored and a private holdout is withheld against training-data contamination.
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 20 July 2026