大型语言模型在情境化医疗问答中的指令调优与思维链提示
Instruction Tuning and CoT Prompting for Contextual Medical QA with LLMs
AI总结：
本研究针对开源大型语言模型在生物医学问答基准PubMedQA上的性能提升问题，通过对比标准指令提示、思维链提示及QLoRA参数高效指令调优策略，发现思维链提示可提升零样本推理，指令调优显著提高准确率，但思维链提示微调的收益具模型和规模依赖性。
摘要：
大型语言模型（LLMs）在医疗问答（MedQA）中展现出巨大潜力，但由于领域特定的复杂性和有限的监督，使其适应生物医学推理仍具挑战。本研究探讨提示设计和轻量级微调如何影响开源LLMs在PubMedQA（生物医学多项选择题基准）上的性能。我们聚焦于两种广泛使用的提示策略——标准指令提示和思维链（CoT）提示，并应用QLoRA进行参数高效的指令调优。在多个模型家族和规模的实验中，我们发现仅CoT提示即可在零样本设置中提升推理能力，而指令调优显著提高准确率。然而，基于CoT提示的微调并非普遍提升性能，甚至可能降低某些大型模型的性能。这些发现表明，推理感知提示是有用的，但其收益取决于模型和规模。我们的研究为医疗问答应用中结合提示工程与高效微调提供了实用见解。
英文摘要：
Large language models (LLMs) have shown great potential in medical question answering (MedQA), yet adapting them to biomedical reasoning remains challenging due to domain-specific complexity and limited supervision. In this work, we study how prompt design and lightweight fine-tuning affect the performance of open-source LLMs on PubMedQA, a benchmark for multiple-choice biomedical questions. We focus on two widely used prompting strategies - standard instruction prompts and Chain-of-Thought (CoT) prompts - and apply QLoRA for parameter-efficient instruction tuning. Across multiple model families and sizes, our experiments show that CoT prompting alone can improve reasoning in zero-shot settings, while instruction tuning significantly boosts accuracy. However, fine-tuning on CoT prompts does not universally enhance performance and may even degrade it for certain larger models. These findings suggest that reasoning-aware prompts are useful, but their benefits are model- and scale-dependent. Our study offers practical insights into combining prompt engineering with efficient finetuning for medical QA applications.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
