Few shot chain-of-thought driven reasoning to prompt LLMs for open ended medical question answering

作者：Ojas Gramopadhye, Saeel Sandeep Nachane, Prateek Chanda, Ganesh Ramakrishnan, K. Jadhav, Yatin Nandwani, Dinesh Raghu, Sachindra Joshi · 发表于：Conference on Empirical Methods in Natural Language Processing · 年份：2024 · DOI：10.48550/arxiv.2403.04890 · 被引用次数：74 · 研究领域：Computer Science

In this paper, we propose a modified version of the MedQA-USMLE dataset, named MEDQA-OPEN, which contains open-ended medical questions without options to mimic clinical scenarios, along with clinician-approved reasoned answers. Additionally, we implement a prompt driven by Chain of Thought (CoT) reasoning, CLINICR, to mirror the prospective process of incremental reasoning, reaching a correct response to medical questions. We empirically demonstrate how CLINICR outperforms the state-of-the-art 5-shot CoT-based prompt (Li\'evin et al., 2022). We also present an approach that mirrors real-life clinical practice by first exploring multiple differential diagnoses through MCQ-CLINICR and subsequently narrowing down to a final diagnosis using MCQ-ELIMINATIVE. Finally, emphasizing the importance of response verification in medical settings, we utilize a reward model mechanism, replacing the elimination process performed by MCQ-ELIMINATIVE.