resource · 2024
IC-BERT: An Instruction Classifier Model Alleviates the Hallucination of Large Language Models in Traditional Chinese Medicine
IEEE International Conference on Computational Intelligence and Applications (IC
IC-BERT: An Instruction Classifier Model Alleviates the Hallucination of Large Language Models in Traditional Chinese Medicine is a resource in the TCM LLM catalog (Awesome-TCM-LLM), dated 2024. It is maintained alongside the GitHub list of Traditional Chinese Medicine large language models.
IC-BERT uses a BERT text filter to limit which questions a traditional Chinese medicine large language model may answer, so as to reduce fabricated facts when questions fall outside its capability.