Selective prediction keeps clinicians' overall accuracy roughly intact but shifts errors toward underdiagnosis (18% more missed) and undertreatment (35% more missed) when the AI abstains.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.HC 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On the Limits of Selective AI Prediction: A Case Study in Clinical Decision Making
Selective prediction keeps clinicians' overall accuracy roughly intact but shifts errors toward underdiagnosis (18% more missed) and undertreatment (35% more missed) when the AI abstains.