Beyond Playtesting: A Generative Multi-Agent Simulation System for Massively Multiplayer Online Games
This provides a cost-efficient framework for data-driven optimization in MMO games, addressing limitations of traditional methods like playtesting, but it is incremental as it adapts existing LLM techniques to a specific domain.
The paper tackles the problem of optimizing numerical systems and mechanism design in Massively Multiplayer Online (MMO) games by proposing a generative agent-based simulation system using Large Language Models (LLMs), which demonstrates strong consistency with real-world player behaviors and plausible causal responses under interventions.
Optimizing numerical systems and mechanism design is crucial for enhancing player experience in Massively Multiplayer Online (MMO) games. Traditional optimization approaches rely on large-scale online experiments or parameter tuning over predefined statistical models, which are costly, time-consuming, and may disrupt player experience. Although simplified offline simulation systems are often adopted as alternatives, their limited fidelity prevents agents from accurately mimicking real player reasoning and reactions to interventions. To address these limitations, we propose a generative agent-based MMO simulation system empowered by Large Language Models (LLMs). By applying Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) on large-scale real player behavioral data, we adapt LLMs from general priors to game-specific domains, enabling realistic and interpretable player decision-making. In parallel, a data-driven environment model trained on real gameplay logs reconstructs dynamic in-game systems. Experiments demonstrate strong consistency with real-world player behaviors and plausible causal responses under interventions, providing a reliable, interpretable, and cost-efficient framework for data-driven numerical design optimization.