Agent improves video games by playing testing and learning from experience
RSIGame: Autonomous Agentic Game Development with Recursive Self-improvement
Computation and LanguageComputer Science and Game TheoryMachine LearningMultiagent Systems
Summary
Making video games automatically with AI is getting better, but fixing bugs and making games fun for many players is still hard. The authors designed RSIGame, an AI system that keeps playing and checking the game to find problems, fix them, and remember what worked well. RSIGame uses detailed step-by-step loops to spot issues and improve game quality over time, even training itself to get better at making games. When tested on many game design tasks, RSIGame consistently made games better than previous methods.
What this means in practice
- •For video game developers: Improve game quality and fix bugs automatically using a self-improving AI agent during development.
- •For software testing teams: Use recursive self-improvement techniques to identify and prioritize software issues beyond initial test coverage.
Authors
Wenyi Wu, Minghao Fu, Jieyu You, Kun Zhou, Siqi Liu, Aayush Salvi, Yiheng Lin, Ce Zhang, Xiaohan Lan, Jiahui Zhu, Yujie Zhong, Qi She, Biwei Huang
Abstract
Recent advances in large language models have made automatic game generation increasingly feasible, yet reliably improving generated games beyond a playable version remains challenging. Naive iterative refinement can easily overfit a small set of test cases, producing fragile games with unresolved bugs, missing behaviors, and poor generalization to broader player interactions. We introduce RSIGame, an autonomous agentic game development framework with recursive self-improvement. RSIGame organizes development into complementary local and global loops. Concretely, a local explore-diagnose-improve loop broadly explores the executable game, diagnoses and prioritizes discovered issues, and performs evidence-grounded revision, where an evolving checklist continually accumulates new testing and improvement guidance. A global loop tracks overall quality, preserves the best checkpoint, and detects saturation or regression over long-horizon development. Beyond test-time improvement, RSIGame further internalizes successful development experience into the generator through training. Across 140 GameCraft-Bench tasks, two game engines, and five generators, RSIGame consistently improves game quality under matched development budgets. Notably, experience internalization enables Qwen3.8-27B to reach 61.38 on Godot and 58.53 on Phaser, exceeding GPT-5.5 one-shot scores while reducing Qwen's generation tokens by 11 times.