September 29, 2024BioengineeringOpen Access

Comparative Analysis of Large Language Models in Chinese Medical Named Entity Recognition

View Full Paper

Ask AI

Bookmark

Authors

ZZZhichao ZhuFudan University QZQing ZhaoTianjin University JLJianjiang LiBeijing University of Technology

Discussion

Member takes

Implication

Comparative evaluation demonstrates enhanced performance of large language models in Chinese biomedical named entity recognition after instruction fine-tuning, highlighting domain adaptation limits.

Key Points

Large language models show promising yet limited Chinese biomedical named entity recognition without targeted instruction fine-tuning.
Fine-tuned ChatGLM2-6B outperformed the BiLSTM CRF baseline, while fine-tuned GPT-3.5 led all language models on the CCKS2017 dataset.
Comparative evaluation examined ChatGLM2-6B, GLM-130B, GPT-3.5, and GPT-4 across a Chinese electronic medical record and CCKS2017 dataset.
Highlights both the potential of instruction fine-tuning and the persistent challenge of outperforming specialized task-specific models.

Cite This Study

Zhu et al. (2024) studied this question.

synapsesocial.com/papers/68e57074b6db6435875114ca https://doi.org/10.3390/bioengineering11100982

View Full Paper

Ask AI

Bookmark

Implication

Comparative evaluation demonstrates enhanced performance of large language models in Chinese biomedical named entity recognition after instruction fine-tuning, highlighting domain adaptation limits.

Comparative Analysis of Large Language Models in Chinese Medical Named Entity Recognition

Discussion

Member takes

Implication

Key Points

Cite This Study

Comparative Analysis of Large Language Models in Chinese Medical Named Entity Recognition

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider