AeroSec: a vertical domain large language model for air traffic management cybersecurity
Ruochen Dong, Chengkai Piao, Buhong Wang, Yongjian Guan, Zhengyang Zhao, Xiangyu Xie · 2026
Read AeroSec: a vertical domain large language model for air traffic management cybersecurity by Ruochen Dong, Chengkai Piao, Buhong Wang, Yongjian Guan, Zhengyang Zhao, Xiangyu Xie (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/PMC13462638.txt.
Description
Ruochen Dong and colleagues build, in a 2026 computing study, a specialized language model for cybersecurity in air traffic management. Air traffic systems link to outside networks, exposing them to attacks such as navigation spoofing and fake aircraft signals. General language models and aviation models focus on passengers and operations, leaving security gaps. The team harvested knowledge from sources and reshaped it into question-and-answer pairs for training.
Two base models were tuned with a mix of full and low-cost methods, using plain instruction tuning and reasoning-focused tuning. A data set and benchmark for the specialty were built alongside the model, named AeroSec. Judged by a third-party model, AeroSec beat its base models and two general models on specialty questions. The authors argue the work fills a blind spot where aviation safety meets defense, and gives the field a shared yardstick.
Catalog record
| CREATOR | Ruochen Dong, Chengkai Piao, Buhong Wang, Yongjian Guan, Zhengyang Zhao, Xiangyu Xie |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44594 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |