This research paper explores the fusion of computational experiments and Large Language Models (LLMs) in studying complex systems. Computational experiments are valuable for studying complex systems, but accurately representing real social systems in Agent-based Modeling (ABM) is challenging due to human characteristics like bounded rationality and heterogeneity. To address this, the integration of LLMs has been proposed, enabling agents to have complex reasoning and autonomous learning abilities. The paper outlines the historical development of agent structures and their evolution into artificial societies, emphasizing their importance in computational experiments. It discusses the advantages that computational experiments and LLM-based Agents offer each other and addresses the challenges and future trends in this research domain.
Publication date: 2 Feb 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2402.00262
