Authoritative Textbook-Augmented Large Language Models for High-Altitude Public Health Medical Education in the Xizang Autonomous Region: Cross-Sectional Comparative Evaluation Study
Kun He, Qiming Xiao, Wangyang Chen, Lisha Jing, Yabing Wang, Shuai Li, Daiyu Yang, Hemiao Xu, Ke Pang, Ruoyu Xiao, Zhashilamu Liu, Deji Zhuoga, Ruxuan Chen, Jingyi Li, Long Chang, Yangzhong Zhou, Zhe Zhang, Ran Li, Lujing Ying, Rutong Li, Hongwei Wang, Xin Yin, Ge Zhen, Siyi Cai, Qijun Shan, Qiang Wang, Danzeng Zhuoga, Ciren Yangjin, Gesang Luobu, Tu Ji, Dong Wu, Stefano Brini · 2026
Read Authoritative Textbook-Augmented Large Language Models for High-Altitude Public Health Medical Education in the Xizang Autonomous Region: Cross-Sectional Comparative Evaluation Study by Kun He, Qiming Xiao, Wangyang Chen, Lisha Jing, Yabing Wang, Shuai Li, Daiyu Yang, Hemiao Xu, Ke Pang, Ruoyu Xiao, Zhashilamu Liu, Deji Zhuoga, Ruxuan Chen, Jingyi Li, Long Chang, Yangzhong Zhou, Zhe Zhang, Ran Li, Lujing Ying, Rutong Li, Hongwei Wang, Xin Yin, Ge Zhen, Siyi Cai, Qijun Shan, Qiang Wang, Danzeng Zhuoga, Ciren Yangjin, Gesang Luobu, Tu Ji, Dong Wu, Stefano Brini (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13271586.txt.
Description
Kun He is lead author of a 2026 cross-sectional comparative study on using large language models, boosted with authoritative textbooks, for public health medical education in Tibet. The region, formally the Xizang Autonomous Region, is high-altitude and low-resource. Old textbooks lack accessibility and interaction, while general AI models can invent facts in specialised medicine.
In the first stage, four public models, from OpenAI, Google, DeepSeek, and Tencent, answered an 80-question benchmark on high-altitude public health, written by specialists. Each question was asked three times, giving 960 outputs, and two blinded panels of eight physicians scored the first replies for completeness, accuracy, clarity, and relevance. A second stage tested adding textbook retrieval to the models.
Catalog record
| CREATOR | Kun He, Qiming Xiao, Wangyang Chen, Lisha Jing, Yabing Wang, Shuai Li, Daiyu Yang, Hemiao Xu, Ke Pang, Ruoyu Xiao, Zhashilamu Liu, Deji Zhuoga, Ruxuan Chen, Jingyi Li, Long Chang, Yangzhong Zhou, Zhe Zhang, Ran Li, Lujing Ying, Rutong Li, Hongwei Wang, Xin Yin, Ge Zhen, Siyi Cai, Qijun Shan, Qiang Wang, Danzeng Zhuoga, Ciren Yangjin, Gesang Luobu, Tu Ji, Dong Wu, Stefano Brini |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55500 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |