resource · 2025
轻量化中医药古籍大语言模型（LoRA）
世界科学技术-中医药现代化
轻量化中医药古籍大语言模型（LoRA） is a resource in the TCM LLM catalog (Awesome-TCM-LLM), dated 2025. It is maintained alongside the GitHub list of Traditional Chinese Medicine large language models.
Shanghan Lun QA pairs from GPT-4, plus ShenNong and cMedQA2, are used to LoRA-tune five base models and to compare full LoRA with 8-bit and 4-bit QLoRA on Qwen-7B-Chat. That model gains 17.61 BLEU and 21.4 ROUGE-L over its base. No separate model name.