MedChatZH: a Better Medical Adviser Learns from Better Instructions
This addresses the need for better medical advice in traditional Chinese medicine, but it is incremental as it applies existing fine-tuning methods to a specialized domain.
The paper tackles the problem of poor performance of large language models in traditional Chinese medical question-answering by introducing MedChatZH, a model pre-trained on medical books and fine-tuned with curated instructions, which outperforms baselines on a real-world dataset.
Generative large language models (LLMs) have shown great success in various applications, including question-answering (QA) and dialogue systems. However, in specialized domains like traditional Chinese medical QA, these models may perform unsatisfactorily without fine-tuning on domain-specific datasets. To address this, we introduce MedChatZH, a dialogue model designed specifically for traditional Chinese medical QA. Our model is pre-trained on Chinese traditional medical books and fine-tuned with a carefully curated medical instruction dataset. It outperforms several solid baselines on a real-world medical dialogue dataset. We release our model, code, and dataset on https://github.com/tyang816/MedChatZH to facilitate further research in the domain of traditional Chinese medicine and LLMs.