Implementation approach of large language model of zero-shot training on collaboration in TCM knowledge reasoning

Key Points

• This research aims to address challenges faced by large language models (LLMs) in reasoning and decision-making within traditional Chinese medicine (TCM).
• Developed a zero-shot training model focusing on multidisciplinary collaboration in TCM.
• Utilized role-based simulations to mimic expert consultations across various fields.
• Emphasized the unique features of TCM in the implementation process.
• Enhanced LLMs' ability to reason clinically in TCM settings.
• Proposed a new framework for expert collaboration that improves diagnostic processes.
• Demonstrated increased efficacy of LLMs in addressing clinical TCM challenges.