Technical report
Grounding Clinical Note Generation Against Transcript Evidence
How reliably can span level grounding detect fabricated clinical content before a clinician sees it?
- Kind
- Technical report
- Year
- 2025
- Status
- In progress
01
Abstract
A working report on the evaluation approach behind Nora Health. Summarisation metrics reward fluency and are close to blind to fabrication, which is the failure that matters in a clinical record. This report sets out an evaluation that decomposes a generated note into individual clinical assertions, requires each to anchor to a transcript span, and reports grounding separately from readability. It includes the error taxonomy that came out of structured review sessions with clinicians.
02
Methods
- BERTScore
- Span alignment
- Error taxonomy
- Human review protocol