Large language models encode clinical knowledge

🤖 Plain-English Summary

Abstract Large language models (LLMs) have demonstrated impressive capabilities, but the bar for clinical applications is high. We show that comprehension, knowledge recall and reasoning improve with model scale and instruction prompt tuning, suggesting the potential utility of LLMs in medicine.

🔑 Key Findings

• Attempts to assess the clinical knowledge of models typically rely on automated evaluations based on limited benchmarks.
• Here, to address these limitations, we present MultiMedQA, a benchmark combining six existing medical question answering datasets spanning professional medicine, research and consumer queries and a new dataset of medical questions searched online, HealthSearchQA.
• We propose a human evaluation framework for model answers along multiple axes including factuality, comprehension, reasoning, possible harm and bias.

🔗 Related Resources