Fly, Fail, Fix: Iterative Game Repair with Reinforcement Learning and Large Multimodal Models

TL;DR

This paper introduces an automated game design iteration framework combining reinforcement learning and large multimodal models to analyze gameplay and iteratively improve game mechanics based on player behavior.

Contribution

It presents a novel iterative framework that uses RL agents and LMMs to analyze gameplay and automatically revise game configurations, advancing AI-assisted game design.

Findings

LMMs can interpret gameplay traces to guide game revisions.

The framework effectively refines game mechanics through iterative playtesting.

Results demonstrate scalable AI tools for dynamic game design.

Abstract

Game design hinges on understanding how static rules and content translate into dynamic player behavior - something modern generative systems that inspect only a game's code or assets struggle to capture. We present an automated design iteration framework that closes this gap by pairing a reinforcement learning (RL) agent, which playtests the game, with a large multimodal model (LMM), which revises the game based on what the agent does. In each loop the RL player completes several episodes, producing (i) numerical play metrics and/or (ii) a compact image strip summarising recent video frames. The LMM designer receives a gameplay goal and the current game configuration, analyses the play traces, and edits the configuration to steer future behaviour toward the goal. We demonstrate results that LMMs can reason over behavioral traces supplied by RL agents to iteratively refine game mechanics, pointing toward practical, scalable tools for AI-assisted game design.