Efficient Medical Question Answering with Knowledge-Augmented Question Generation
This work addresses the problem of specialized medical knowledge representation for small language models, offering an incremental improvement in domain-specific applications.
The paper tackled the challenge of improving small language models for medical question answering by fine-tuning them on medical textbooks and using GPT-4 to generate knowledge-augmented questions, resulting in enhanced proficiency on the novel ECN-QA dataset.
In the expanding field of language model applications, medical knowledge representation remains a significant challenge due to the specialized nature of the domain. Large language models, such as GPT-4, obtain reasonable scores on medical question answering tasks, but smaller models are far behind. In this work, we introduce a method to improve the proficiency of a small language model in the medical domain by employing a two-fold approach. We first fine-tune the model on a corpus of medical textbooks. Then, we use GPT-4 to generate questions similar to the downstream task, prompted with textbook knowledge, and use them to fine-tune the model. Additionally, we introduce ECN-QA, a novel medical question answering dataset containing ``progressive questions'' composed of related sequential questions. We show the benefits of our training strategy on this dataset. The study's findings highlight the potential of small language models in the medical domain when appropriately fine-tuned. The code and weights are available at https://github.com/raidium-med/MQG.