A fact-alignment LLM method detects clinical summarization hallucinations better than existing metrics, with correlations of 0.43 on controlled data and 0.37 on natural errors.
Addi- tionally, our research focuses solely on SOAP notes, excluding other clinical documents like discharge summaries or progress reports
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Fact-Controlled Diagnosis of Hallucinations in Medical Text Summarization
A fact-alignment LLM method detects clinical summarization hallucinations better than existing metrics, with correlations of 0.43 on controlled data and 0.37 on natural errors.