ScriptDoctor: Automatic Generation of PuzzleScript Games via Large Language Models and Tree Search
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912418169356288 |
|---|---|
| author | Earle, Sam Khalifa, Ahmed Nasir, Muhammad Umair Jiang, Zehua Todd, Graham Banburski-Fahey, Andrzej Togelius, Julian |
| author_facet | Earle, Sam Khalifa, Ahmed Nasir, Muhammad Umair Jiang, Zehua Todd, Graham Banburski-Fahey, Andrzej Togelius, Julian |
| contents | There is much interest in using large pre-trained models in Automatic Game Design (AGD), whether via the generation of code, assets, or more abstract conceptualization of design ideas. But so far this interest largely stems from the ad hoc use of such generative models under persistent human supervision. Much work remains to show how these tools can be integrated into longer-time-horizon AGD pipelines, in which systems interface with game engines to test generated content autonomously. To this end, we introduce ScriptDoctor, a Large Language Model (LLM)-driven system for automatically generating and testing games in PuzzleScript, an expressive but highly constrained description language for turn-based puzzle games over 2D gridworlds. ScriptDoctor generates and tests game design ideas in an iterative loop, where human-authored examples are used to ground the system's output, compilation errors from the PuzzleScript engine are used to elicit functional code, and search-based agents play-test generated games. ScriptDoctor serves as a concrete example of the potential of automated, open-ended LLM-based workflows in generating novel game content. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_06524 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ScriptDoctor: Automatic Generation of PuzzleScript Games via Large Language Models and Tree Search Earle, Sam Khalifa, Ahmed Nasir, Muhammad Umair Jiang, Zehua Todd, Graham Banburski-Fahey, Andrzej Togelius, Julian Artificial Intelligence Human-Computer Interaction There is much interest in using large pre-trained models in Automatic Game Design (AGD), whether via the generation of code, assets, or more abstract conceptualization of design ideas. But so far this interest largely stems from the ad hoc use of such generative models under persistent human supervision. Much work remains to show how these tools can be integrated into longer-time-horizon AGD pipelines, in which systems interface with game engines to test generated content autonomously. To this end, we introduce ScriptDoctor, a Large Language Model (LLM)-driven system for automatically generating and testing games in PuzzleScript, an expressive but highly constrained description language for turn-based puzzle games over 2D gridworlds. ScriptDoctor generates and tests game design ideas in an iterative loop, where human-authored examples are used to ground the system's output, compilation errors from the PuzzleScript engine are used to elicit functional code, and search-based agents play-test generated games. ScriptDoctor serves as a concrete example of the potential of automated, open-ended LLM-based workflows in generating novel game content. |
| title | ScriptDoctor: Automatic Generation of PuzzleScript Games via Large Language Models and Tree Search |
| topic | Artificial Intelligence Human-Computer Interaction |
| url | https://arxiv.org/abs/2506.06524 |