IPCGRL: Language-Instructed Reinforcement Learning for Procedural Level Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Baek, In-Chang, Kim, Sung-Hyun, Lee, Seo-Young, Kim, Dong-Hyeon, Kim, Kyung-Joong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909702957301760
author Baek, In-Chang
Kim, Sung-Hyun
Lee, Seo-Young
Kim, Dong-Hyeon
Kim, Kyung-Joong
author_facet Baek, In-Chang
Kim, Sung-Hyun
Lee, Seo-Young
Kim, Dong-Hyeon
Kim, Kyung-Joong
contents Recent research has highlighted the significance of natural language in enhancing the controllability of generative models. While various efforts have been made to leverage natural language for content generation, research on deep reinforcement learning (DRL) agents utilizing text-based instructions for procedural content generation remains limited. In this paper, we propose IPCGRL, an instruction-based procedural content generation method via reinforcement learning, which incorporates a sentence embedding model. IPCGRL fine-tunes task-specific embedding representations to effectively compress game-level conditions. We evaluate IPCGRL in a two-dimensional level generation task and compare its performance with a general-purpose embedding method. The results indicate that IPCGRL achieves up to a 21.4% improvement in controllability and a 17.2% improvement in generalizability for unseen instructions. Furthermore, the proposed method extends the modality of conditional input, enabling a more flexible and expressive interaction framework for procedural content generation.
format Preprint
id arxiv_https___arxiv_org_abs_2503_12358
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle IPCGRL: Language-Instructed Reinforcement Learning for Procedural Level Generation
Baek, In-Chang
Kim, Sung-Hyun
Lee, Seo-Young
Kim, Dong-Hyeon
Kim, Kyung-Joong
Artificial Intelligence
Computation and Language
Machine Learning
Recent research has highlighted the significance of natural language in enhancing the controllability of generative models. While various efforts have been made to leverage natural language for content generation, research on deep reinforcement learning (DRL) agents utilizing text-based instructions for procedural content generation remains limited. In this paper, we propose IPCGRL, an instruction-based procedural content generation method via reinforcement learning, which incorporates a sentence embedding model. IPCGRL fine-tunes task-specific embedding representations to effectively compress game-level conditions. We evaluate IPCGRL in a two-dimensional level generation task and compare its performance with a general-purpose embedding method. The results indicate that IPCGRL achieves up to a 21.4% improvement in controllability and a 17.2% improvement in generalizability for unseen instructions. Furthermore, the proposed method extends the modality of conditional input, enabling a more flexible and expressive interaction framework for procedural content generation.
title IPCGRL: Language-Instructed Reinforcement Learning for Procedural Level Generation
topic Artificial Intelligence
Computation and Language
Machine Learning
url https://arxiv.org/abs/2503.12358