A systematic study of 2800 AI responses shows automated metrics using clinician references closely match human ratings for answering hospitalization questions, using clinical notes, and applying medical knowledge.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Automated Evaluation can Distinguish the Good and Bad AI Responses to Patient Questions about Hospitalization
A systematic study of 2800 AI responses shows automated metrics using clinician references closely match human ratings for answering hospitalization questions, using clinical notes, and applying medical knowledge.