Can large language models reason about medical questions?

Research output: Contribution to journalJournal articleResearchpeer-review


  • Fulltext

    Final published version, 3.72 MB, PDF document

  • Valentin Liévin
  • Christoffer Egeberg Hother
  • Andreas Geert Motzfeldt
  • Winther, Ole

Although large language models often produce impressive outputs, it remains unclear how they perform in real-world scenarios requiring strong reasoning skills and expert domain knowledge. We set out to investigate whether closed- and open-source models (GPT-3.5, Llama 2, etc.) can be applied to answer and reason about difficult real-world-based questions. We focus on three popular medical benchmarks (MedQA-US Medical Licensing Examination [USMLE], MedMCQA, and PubMedQA) and multiple prompting scenarios: chain of thought (CoT; think step by step), few shot, and retrieval augmentation. Based on an expert annotation of the generated CoTs, we found that InstructGPT can often read, reason, and recall expert knowledge. Last, by leveraging advances in prompt engineering (few-shot and ensemble methods), we demonstrated that GPT-3.5 not only yields calibrated predictive distributions but also reaches the passing score on three datasets: MedQA-USMLE (60.2%), MedMCQA (62.7%), and PubMedQA (78.2%). Open-source models are closing the gap: Llama 2 70B also passed the MedQA-USMLE with 62.5% accuracy.

Original languageEnglish
Article number100943
Issue number3
Number of pages12
Publication statusPublished - 2024

Bibliographical note

Publisher Copyright:
© 2024 The Authors

    Research areas

  • GPT-3.5, large language models, Llama 2, machine learning, medical, MedQA, open source, prompt engineering, question answering, uncertainty quantification

ID: 385123211