Utilizing Large language models to select literature for meta-analysis shows workload reduction while maintaining a similar recall level as manual curation
We show here the groundbreaking finding that automatic selection of literature for meta-analysis is possible with LLMs. We provide it here as a pipeline, LARS-GPT, which showed a great workload reduction while maintaining a pre-set recall.
Authors
- Juan J. García‐Vallejo (ORCID: https://orcid.org/0000-0001-6238-7069)
- Yuanming Geng
- Duolao Wang (ORCID: https://orcid.org/0000-0003-2788-2464)
- Chiyuan Ma (ORCID: https://orcid.org/0000-0002-1710-5955)
- Bart A. Westerman (ORCID: https://orcid.org/0000-0002-9898-9616)
- Yiming Du
- Xiangming Cai (ORCID: https://orcid.org/0000-0002-3861-664X)
Institutions
- Chinese University of Hong Kong (HK)
- Nanjing General Hospital of Nanjing Military Command (CN)
- Amsterdam Neuroscience (NL)
- Amsterdam University Medical Centers (NL)
- Jinling Institute of Technology (CN)
- Nanjing Medical University (CN)
- Vrije Universiteit Amsterdam (NL)
Publication Details
- Journal
- BMC Medical Research Methodology
- Published
- 2025-04-28
- DOI
- https://doi.org/10.1186/s12874-025-02569-3
- Citations
- 9
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 5.89
Funders
- China Scholarship Council