Taiyi: a bilingual fine-tuned large language model for diverse biomedical tasks
Journal of the American Medical Informatics Association · 01.09.2024 · 31 (9) · 1865-1874
DOI: 10.1093/jamia/ocae037 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, online erschienen am 29.02.2024. Verfasst von Ling Luo und 19 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, China. Hauptthemen laut MEDLINE: Natural Language Processing, Data Mining, Multilingualism. Die Arbeit wurde bisher 81-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities.
Abstract (Originalsprache)
Objective
Most existing fine-tuned biomedical large language models (LLMs) focus on enhancing performance in monolingual biomedical question answering and conversation tasks. To investigate the effectiveness of the fine-tuned LLMs on diverse biomedical natural language processing (NLP) tasks in different languages, we present Taiyi, a bilingual fine-tuned LLM for diverse biomedical NLP tasks. Materials and Methods We first curated a comprehensive collection of 140 existing biomedical text mining datasets (102 English and 38 Chinese datasets) across over 10 task types. Subsequently, these corpora were converted to the instruction data used to fine-tune the general LLM. During the supervised fine-tuning phase, a 2-stage strategy is proposed to optimize the model performance across various tasks.
Results
Experimental results on 13 test sets, which include named entity recognition, relation extraction, text classification, and question answering tasks, demonstrate that Taiyi achieves superior performance compared to general LLMs. The case study involving additional biomedical NLP tasks further shows Taiyi’s considerable potential for bilingual biomedical multitasking.
Conclusion
Leveraging rich high-quality biomedical corpora and developing effective fine-tuning strategies can significantly improve the performance of LLMs within the biomedical domain. Taiyi shows the bilingual multitasking capability through supervised fine-tuning. However, those tasks such as information extraction that are not generation tasks in nature remain challenging for LLM-based generative approaches, and they still underperform the conventional discriminative approaches using smaller language models.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, China.
ORCID-Profile: Ling Luo, Peng Chen, Zhihao Yang
Förderung
- National Natural Science Foundation of China
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Luo, L., Ning, J., Zhao, Y., Wang, Z., Ding, Z., Chen, P., Fu, W., Han, Q., Xu, G., Qiu, Y., Pan, D., Li, J., Li, H., Feng, W., Tu, S., Liu, Y., Yang, Z., Wang, J., Sun, Y., & Lin, H. (2024). Taiyi: a bilingual fine-tuned large language model for diverse biomedical tasks. Journal of the American Medical Informatics Association, 31(9), 1865-1874. https://doi.org/10.1093/jamia/ocae037
Vancouver
Luo L, Ning J, Zhao Y, Wang Z, Ding Z, Chen P, et al. Taiyi: a bilingual fine-tuned large language model for diverse biomedical tasks. Journal of the American Medical Informatics Association. 2024;31(9):1865-1874. doi:10.1093/jamia/ocae037
BibTeX
@article{Luo2024,
author = {Luo, Ling and Ning, Jinzhong and Zhao, Yingwen and Wang, Zhijun and Ding, Zeyuan and Chen, Peng and Fu, Weiru and Han, Qinyu and Xu, Guangtao and Qiu, Yunzhi and Pan, Dinghao and Li, Jiru and Li, Hao and Feng, Wenduo and Tu, Senbo and Liu, Yuqi and Yang, Zhihao and Wang, Jian and Sun, Yuanyuan and Lin, Hongfei},
title = {Taiyi: a bilingual fine-tuned large language model for diverse biomedical tasks},
journal = {Journal of the American Medical Informatics Association},
year = {2024},
volume = {31},
number = {9},
pages = {1865--1874},
doi = {10.1093/jamia/ocae037}
}
EUVIMED