EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Triospect: A Three-Dimensional Framework for Robust Statistical AI-Generated Text Detection Against Diverse Attacks

Guangsheng Bao, Lihua Rong, Yanbin Zhao, Xiao Yu, Qiji Zhou, Yue Zhang

Transactions of the Association for Computational Linguistics · 05.11.2026 · 14 · 2775-2793

DOI: 10.1162/tacl.a.825 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 05.11.2026. Verfasst von Guangsheng Bao und 5 weiteren Autor:innen. Die Arbeit nennt 83 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Existing AI-generated text detectors are vulnerable to attacks that manipulate textual characteristics. In this study, we propose a novel Triospect Detection Framework by using additional perspectives of content (core ideas) and expression (stylistic elements) within a given text. Experiments on two benchmarks involving 17 attacks, 12 domains, and 17 source models demonstrate that Triospect is robust against these attacks. It improves the strong baseline by a significant margin of 22.3% (AUROC) and 13% (TPR01) on the Humanize-16K after-attack subset, and by 9.1% (AUROC) and 22% (TPR01) on the adversarial RAID. This framework marks a pioneering effort in statistical methods to enhance detection reliability against attacks. We release our data and code at https://github.com/baoguangsheng/triospect.

Zitieren

APA

Bao, G., Rong, L., Zhao, Y., Yu, X., Zhou, Q., & Zhang, Y. (2026). Triospect: A Three-Dimensional Framework for Robust Statistical AI-Generated Text Detection Against Diverse Attacks. Transactions of the Association for Computational Linguistics, 14, 2775-2793. https://doi.org/10.1162/tacl.a.825

Vancouver

Bao G, Rong L, Zhao Y, Yu X, Zhou Q, Zhang Y. Triospect: A Three-Dimensional Framework for Robust Statistical AI-Generated Text Detection Against Diverse Attacks. Transactions of the Association for Computational Linguistics. 2026;14:2775-2793. doi:10.1162/tacl.a.825

BibTeX
@article{Bao2026,
  author = {Bao, Guangsheng and Rong, Lihua and Zhao, Yanbin and Yu, Xiao and Zhou, Qiji and Zhang, Yue},
  title = {Triospect: A Three-Dimensional Framework for Robust Statistical AI-Generated Text Detection Against Diverse Attacks},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2026},
  volume = {14},
  pages = {2775--2793},
  doi = {10.1162/tacl.a.825}
}
Kostenlos starten →