IvyGPT: InteractiVe Chinese pathwaY language model in medical domain
This work addresses the need for accurate and interactive medical language models in the Chinese healthcare domain, representing an incremental improvement over existing medical GPT models.
The authors tackled the problem of poor accuracy and inability to provide medical advice in general large language models by proposing IvyGPT, a Chinese medical LLM based on LLaMA, trained with high-quality QA instances and RLHF, which outperformed other medical GPT models in experiments.
General large language models (LLMs) such as ChatGPT have shown remarkable success. However, such LLMs have not been widely adopted for medical purposes, due to poor accuracy and inability to provide medical advice. We propose IvyGPT, an LLM based on LLaMA that is trained and fine-tuned with high-quality medical question-answer (QA) instances and Reinforcement Learning from Human Feedback (RLHF). After supervised fine-tuning, IvyGPT has good multi-turn conversation capabilities, but it cannot perform like a doctor in other aspects, such as comprehensive diagnosis. Through RLHF, IvyGPT can output richer diagnosis and treatment answers that are closer to human. In the training, we used QLoRA to train 33 billion parameters on a small number of NVIDIA A100 (80GB) GPUs. Experimental results show that IvyGPT has outperformed other medical GPT models.