Large Language Models Need Consultants for Reasoning: Becoming an Expert in a Complex Human System through Behavioral Simulation
ACM Transactions on Intelligent Systems and Technology · 31.10.2026 · 17 (5) · 1-24
DOI: 10.1145/3800956 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Intelligent Systems and Technology, online erschienen am 23.07.2026. Verfasst von Chuwen Wang und 2 weiteren Autor:innen. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shanghai Municipal Commission of Economy and Informatization, National Natural Science Foundation of China, Key R&D Program of Shandong Province, China.
Abstract (Originalsprache)
Large Language Models (LLMs), in conjunction with various reasoning reinforcement methodologies, have demonstrated remarkable capabilities comparable to humans in fields such as mathematics, law, coding, common sense, and world knowledge. In this article, we improve the reasoning abilities of LLMs within complex human systems. We propose a novel reasoning framework, termed “Mosaic Expert Observation Wall” (MEOW) exploiting generative-agent-based simulation technique. In the MEOW framework, simulated data are utilized to train an expert model learning “experience” about a specific task in each independent time of simulation. It is the accumulated “experience” through the simulation that makes for an expert on a task in a complex human system. We conduct the experiments within two communication games that represent two kinds of complex human systems. The results indicate that our proposed methodology can cooperate with existing methodologies to enhance the reasoning abilities of LLMs in complex human systems.
Förderung
- Shanghai Municipal Commission of Economy and Informatization
- National Natural Science Foundation of China
- Key R&D Program of Shandong Province, China
- Fundamental Research Funds for the Central Universities
- Key Laboratory of Computing Power Network and Information Security, Ministry of Education
- Leadership Project under the Eastern Talent Program
Zitieren
APA
Wang, C., Zeng, S., & Wang, C. (2026). Large Language Models Need Consultants for Reasoning: Becoming an Expert in a Complex Human System through Behavioral Simulation. ACM Transactions on Intelligent Systems and Technology, 17(5), 1-24. https://doi.org/10.1145/3800956
Vancouver
Wang C, Zeng S, Wang C. Large Language Models Need Consultants for Reasoning: Becoming an Expert in a Complex Human System through Behavioral Simulation. ACM Transactions on Intelligent Systems and Technology. 2026;17(5):1-24. doi:10.1145/3800956
BibTeX
@article{Wang2026,
author = {Wang, Chuwen and Zeng, Shirong and Wang, Cheng},
title = {Large Language Models Need Consultants for Reasoning: Becoming an Expert in a Complex Human System through Behavioral Simulation},
journal = {ACM Transactions on Intelligent Systems and Technology},
year = {2026},
volume = {17},
number = {5},
pages = {1--24},
doi = {10.1145/3800956}
}
EUVIMED