CrowdLLM: Building LLM-Based Digital Populations Augmented with Generative Models
INFORMS Journal on Data Science · 05.08.2026
DOI: 10.1287/ijds.2025.0140 ↗Kurz gefasst
Fachartikel in der Zeitschrift INFORMS Journal on Data Science, online erschienen am 05.08.2026. Verfasst von Ryan Feng Lin und 5 weiteren Autor:innen. Die Arbeit nennt 112 Literaturangaben.
Abstract (Originalsprache)
The emergence of large language models (LLMs) has sparked much interest in creating LLM-based digital populations that can be applied to many applications such as social simulation, crowdsourcing, marketing, and recommendation systems. A digital population can reduce the cost of recruiting human participants and alleviate many concerns related to human subject study. However, research has found that most of the existing works rely solely on LLMs and could not sufficiently capture the accuracy and diversity of a real human population. To address this limitation, we propose CrowdLLM, which integrates pretrained LLMs and generative models to enhance the diversity and fidelity of the digital population. We conduct theoretical analysis of CrowdLLM regarding its great potential in creating cost-effective, sufficiently representative, scalable digital populations that can match the quality of a real crowd. Comprehensive experiments are also conducted across multiple domains (e.g., crowdsourcing, voting, user rating) and simulation studies which demonstrate that CrowdLLM achieves promising performance in both accuracy and distributional fidelity to human data. History: Yu Ding served as the senior editor for this article. Supplemental Material: The online appendix and code files are available at https://doi.org/10.1287/ijds.2025.0140 .
Zitieren
APA
Lin, R. F., Tian, K., Zheng, H., Zhang, C., Zeng, L., & Huang, S. (2026). CrowdLLM: Building LLM-Based Digital Populations Augmented with Generative Models. INFORMS Journal on Data Science. https://doi.org/10.1287/ijds.2025.0140
Vancouver
Lin RF, Tian K, Zheng H, Zhang C, Zeng L, Huang S. CrowdLLM: Building LLM-Based Digital Populations Augmented with Generative Models. INFORMS Journal on Data Science. 2026. doi:10.1287/ijds.2025.0140
BibTeX
@article{Lin2026,
author = {Lin, Ryan Feng and Tian, Keyu and Zheng, Hanming and Zhang, Congjing and Zeng, Li and Huang, Shuai},
title = {CrowdLLM: Building LLM-Based Digital Populations Augmented with Generative Models},
journal = {INFORMS Journal on Data Science},
year = {2026},
doi = {10.1287/ijds.2025.0140}
}
EUVIMED