Grammar and Gameplay-aligned RL for Game Description Generation with LLMs

Tsunehiko Tanaka, Edgar Simo-Serra
3/20/2025
cs.CLcs.AI

Official code: tsunehiko/rlgdg (1 star, MIT, last commit 2025-07-24).

Abstract

Game Description Generation (GDG) is the task of generating a game description written in a Game Description Language (GDL) from natural language text. Previous studies have explored generation methods leveraging the contextual understanding capabilities of Large Language Models (LLMs); however, accurately reproducing the game features of the game descriptions remains a challenge. In this paper, we propose reinforcement learning-based fine-tuning of LLMs for GDG (RLGDG). Our training method simultaneously improves grammatical correctness and fidelity to game concepts by introducing both grammar rewards and concept rewards. Furthermore, we adopt a two-stage training strategy where Reinforcement Learning (RL) is applied following Supervised Fine-Tuning (SFT). Experimental results demonstrate that our proposed method significantly outperforms baseline methods using SFT alone. Our code is available at https://github.com/tsunehiko/rlgdg

View on arXivView PDF

Code Implementations(1)

tsunehiko/rlgdgOfficial100%
MIT

Grammar and Gameplay-aligned RL for Game Description Generation with LLMs (Conference on Games 2025)

10Java, ShellJun 29, 20251 years agoMIT

Similar Papers

IPCGRL: Language-Instructed Reinforcement Learning for Procedural Level Generation

85% match2 repos12025

From Gameplay Traces to Game Mechanics: Causal Induction with Large Language Models

84% match2026

High-quality generation of dynamic game content via small language models: A proof of concept

83% match2026

PCGRLLM: Large Language Model-Driven Reward Design for Procedural Content Generation Reinforcement Learning

83% match1 repo32025

Language Model-In-The-Loop: Data Optimal Approach to Learn-To-Recommend Actions in Text Games

83% match2023

Play to Generalize: Learning to Reason Through Game Play

83% match1 repo732025

Cite this paper

@article{tanaka2025grammar,
  title  = {Grammar and Gameplay-aligned RL for Game Description Generation with LLMs},
  author = {Tsunehiko Tanaka and Edgar Simo-Serra},
  year   = {2025},
  eprint = {2503.15783},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2503.15783v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2503.15783.svg)](https://codeofpaper.com/paper/2503.15783)

Discussion