Datamart-Agent: LLM-Driven Game-Theoretic Agent for Data Marketplace Modeling

Data marketplaces analyze strategic data exchange among users, platforms, and buyers. However, most existing studies model on static equilibria and complete information, which limits their realism. In this work, we study whether large language model (LLM)-driven agents can make equilibrium-consistent decisions in analytically tractable data marketplaces with evolving and incomplete-information. Specifically, we introduce EvoDM, an agent-based modeling framework that extends the classical static data marketplace to dynamic and incompleteinformation settings while providing tractable equilibrium benchmarks for evaluating agent decisions. Building upon EvoDM, we propose Datamart-Agent, an LLM-driven gametheoretic agent that improves equilibriumconsistent decision execution through dynamic game tree memory and mechanism-guided reflection, without requiring parameter updates. Experiments demonstrate that Datamart-Agent closely matches equilibrium-consistent decision-making, achieving the lowest utility gap and over 20% higher Pass@ϵ than strong baselines. After validating its effectiveness, we employ EvoDM with Datamart-Agent to analyze competition and regulation in assumption-relaxed settings where closedform ground truth is unavailable, providing exploratory simulation-based insights into market dynamics and regulatory effects.