Do LLMs Triage Like Clinicians? A Dynamic Study of Outpatient Referral
A new study published on arXiv explores the effectiveness of Large Language Models (LLMs) in outpatient referral processes. Researchers found that while LLMs do not significantly outperform traditional classifiers in static referral accuracy, they excel in dynamic, multi-turn dialogue scenarios. This is attributed to their ability to ask targeted follow-up questions that effectively reduce uncertainty and aid clinical decision-making. AI
IMPACT LLMs can enhance clinical decision-making by improving interactive diagnostic processes, moving beyond static classification.