Enhancing Decision-Making with Large Language Models through Multi-Agent Fictitious Play

DOI10.48550/arXiv.2606.19308MaRDI QIDQ7267648arXivFDO

Authors Xiaoyan Zhao, Tat Seng Chua, Leyang Shen, Yang Zhang, Chun Kai Ling, Tat-Seng Chua

Publication date 17 June 2026

Published in arXiv

Tldr This paper proposes Multi-Agent Fictitious Play enhancing large language model decision-making by addressing stance entanglement where stakeholder choices remain mutually dependent within multi-agent systems. Leveraging game-theoretic principles to iteratively refine strategies through equilibrium-seeking processes, the method outperforms baselines on tournament strength and robustness in competitive scenarios.

Full work available at URL https://doi.org/10.48550/arXiv.2606.19308

arXiv classification

cs.CLcs.MA

This page was built for publication: Enhancing Decision-Making with Large Language Models through Multi-Agent Fictitious Play

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7267648)