From Preference to Reciprocity: Decentralized Matching with Empirically Grounded LLM-agent Based Modeling

2026-09-28

AI Summary

This study addresses the limitation of traditional matching mechanisms that rely on complete preferences and centralized computation, rendering them ill-suited for decentralized scenarios. To overcome this, we propose a dynamic bipartite matching framework integrating large language model agents with contextual bandits. Using the Chinese marriage market as a simulation environment, our approach decouples subjective preferences from reciprocal decision-making and employs a Logistic-UCB algorithm to achieve decentralized, asynchronous matching without requiring global rankings. Experimental results demonstrate that at a 50×50 scale, the proposed framework yields optimal average reciprocal welfare, exhibits smaller gender ranking disparities, and produces the fewest blocking pairs compared to classical baselines. These findings indicate significant performance improvements over existing methods in decentralized matching settings.

Abstract

Bipartite matching is a fundamental problem in game theory and market design. Classical approaches such as Gale--Shapley assume complete preferences and centralized computation, whereas many real-world matching processes are decentralized, asynchronous, and shaped by sequential interaction under limited information. We propose a dynamic bipartite matching framework that combines large language model (LLM) agents with contextual bandits. In a simulated Chinese marriage market, economically grounded LLM agents evaluate locally encountered candidates, while agent-specific Logistic-UCB models learn reciprocal acceptance from realized proposal outcomes. The mechanism therefore separates two decisions---\emph{whom do I like?} and \emph{who is likely to like me back?}---without requiring ex ante market-wide preference rankings. We first validate LLM-induced mate preferences against the empirical conditional-logit reference across multiple LLM backbones. In the $50\times50$ matching experiment, Bandit-UCB achieves the highest mean mutual welfare (56.01 versus 54.87 for Gale--Shapley), a smaller gender rank gap than the classical baselines, and the fewest blocking pairs among the LLM-ABM policies. Learned acceptance models show economically interpretable gender-differentiated associations, while counterfactual setups reveal no systematic unilateral advantage from prior search knowledge. Overall, these results support the advantages of decentralized matching with LLM-based behavioral modeling and online learning under incomplete information for economic simulation and computational social science research.

Problem

• Bipartite matching
• Decentralized matching
• Incomplete information
• Market design
• LLM-agent based modeling

Innovation

• Decentralized Matching
• LLM Agents
• Contextual Bandits
• Logistic-UCB
• Agent-Based Modeling