EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automated literature research and review-generation method based on large language models

Shican Wu, Xiao Ma, Dehui Luo, Lulu Li, Xiangcheng Shi, Xin Chang, Xiaoyun Lin, Ran Luo, Chunlei Pei, Changying Du, Zhi-Jian Zhao, Jinlong Gong iD

National Science Review · 05.05.2025 · 12 (6)

DOI: 10.1093/nsr/nwaf169 ↗

Kurz gefasst

Fachartikel in der Zeitschrift National Science Review, online erschienen am 25.04.2025. Verfasst von Shican Wu und 11 weiteren Autor:innen. Beteiligte Einrichtung: School of Chemical Engineering and Technology, Key Laboratory for Green Chemical Technology of Ministry of Education, Tianjin University, Collaborati…. Die Arbeit wurde bisher 39-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Program of Introducing Talents of Discipline to Universities.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Literature research, which is vital for scientific work, faces the challenge of surging information volumes that are exceeding researchers’ processing capabilities. This paper describes an automated review-generation method based on large language models (LLMs) to overcome efficiency bottlenecks and reduce cognitive load. Our statistically validated evaluation framework demonstrates that the generated reviews match or exceed manual quality, offering broad applicability across research fields without requiring user domain knowledge. Applied to propane dehydrogenation catalysts, our method demonstrated two aspects: first, generating comprehensive reviews from 343 articles spanning 35 topics; and, second, evaluating data-mining capabilities by using 1041 articles for experimental catalyst property analysis. Through multilayered quality control, we effectively mitigated the hallucinations of LLMs, with expert verification confirming accuracy and citation integrity, while demonstrating hallucination risks reduced to <0.5% with 95% confidence. The released software application enables one-click review generation, enhancing research productivity and literature-recommendation efficiency while facilitating broader scientific explorations.

Schlagwörter

Stichwörter der Autor:innen

automated review generationlarge language modelsliterature analysisscientific writing

Autor:innen und Institutionen

  • School of Chemical Engineering and Technology, Key Laboratory for Green Chemical Technology of Ministry of Education, Tianjin University, Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Ti…
  • Joint School of National University of Singapore and Tianjin University, Fuzhou 350207, China.
  • Zhejiang Institute of Tianjin University, Ningbo 315201, China.
  • International Joint Laboratory of Low-carbon Chemical Engineering of Ministry of Education, Tianjin 300350, China.
  • Haihe Laboratory of Sustainable Chemical Transformations, Tianjin 300192, China.
  • National Industry-Education Platform of Energy Storage, Tianjin University, Tianjin 300350, China.

ORCID-Profile: Jinlong Gong

Förderung

  • National Natural Science Foundation of China
  • Program of Introducing Talents of Discipline to Universities

Zitieren

APA

Wu, S., Ma, X., Luo, D., Li, L., Shi, X., Chang, X., Lin, X., Luo, R., Pei, C., Du, C., Zhao, Z. J., & Gong, J. (2025). Automated literature research and review-generation method based on large language models. National Science Review, 12(6). https://doi.org/10.1093/nsr/nwaf169

Vancouver

Wu S, Ma X, Luo D, Li L, Shi X, Chang X, et al. Automated literature research and review-generation method based on large language models. National Science Review. 2025;12(6). doi:10.1093/nsr/nwaf169

BibTeX
@article{Wu2025,
  author = {Wu, Shican and Ma, Xiao and Luo, Dehui and Li, Lulu and Shi, Xiangcheng and Chang, Xin and Lin, Xiaoyun and Luo, Ran and Pei, Chunlei and Du, Changying and Zhao, Zhi-Jian and Gong, Jinlong},
  title = {Automated literature research and review-generation method based on large language models},
  journal = {National Science Review},
  year = {2025},
  volume = {12},
  number = {6},
  doi = {10.1093/nsr/nwaf169}
}
Kostenlos starten →