Human-Governed Multi-Agent Clinical Reasoning Framework
Large language models (LLMs) show promise in medical reasoning, but their direct deployment in clinical settings faces challenges related to opacity, output instability, and lack of explicit uncertainty control. We present a Human-Governed Multi-Agent Clinical Reasoning Framework that combines LLM-based diagnosis and q...