MeDxAgent: Multi-Agent Consultation for Interactive Medical Diagnosis

Large language models (LLMs) are increasingly used for health-related decision support. Yet most evaluations treat diagnosis as a single-shot task with complete information provided upfront, often as a multiple-choice selection. This diverges from clinical practice, where diagnosis is interactive and open-ended, involving sequential hypothesis refinement through targeted questioning. We address this gap. We build MeDxBench, a large-scale benchmark of $4{,}421$ clinical cases across $20$ specialties. We further propose MeDxAgent, a multi-agent consultation system for interactive diagnosis, and systematically study its prompt-, flow- and agent-level design choices. MeDxAgent achieves a $10.3$% accuracy gain over the baseline on MeDxBench, bridging $52.3$% of the accuracy gap between the baseline and the full-information oracle. We find that specific design choices: collecting demographics first, passing summarized dialogue for diagnosis, and feeding candidate diagnoses for targeted questioning, improve diagnostic accuracy. MeDxAgent also surpasses recent LLM-based diagnostic frameworks, outperforming MedAgentSim and VivaBench by $6.2$% and $20.2$% on their respective benchmarks. Code and dataset will be released alongside the publication.