STORY2GAME: Generating (Almost) Everything in an Interactive Fiction Game

2025-05-06

AI Summary

This work addresses the challenge of generating executable, state-consistent interactive fiction games using large language models (LLMs). Methodologically, it introduces a story-driven end-to-end framework: (1) an LLM first generates a narrative backbone; (2) a dynamic world state representation is constructed from this story; and (3) executable Python action code is automatically synthesized based on formalized action preconditions and effects, with runtime extensibility to accommodate player improvisation. Its key contribution is the first realization of a closed-loop, tightly coupled “story–state–action” generation mechanism, ensuring both narrative openness and interactive executability. Experiments demonstrate a significant improvement in action code generation success rates and enable players to complete generated games end-to-end, thereby validating the feasibility of LLM-driven, fully automated interactive fiction generation.

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

We introduce STORY2GAME, a novel approach to using Large Language Models to generate text-based interactive fiction games that starts by generating a story, populates the world, and builds the code for actions in a game engine that enables the story to play out interactively. Whereas a given set of hard-coded actions can artificially constrain story generation, the ability to generate actions means the story generation process can be more open-ended but still allow for experiences that are grounded in a game state. The key to successful action generation is to use LLM-generated preconditions and effects of actions in the stories as guides for what aspects of the game state must be tracked and changed by the game engine when a player performs an action. We also introduce a technique for dynamically generating new actions to accommodate the player's desire to perform actions that they think of that are not part of the story. Dynamic action generation may require on-the-fly updates to the game engine's state representation and revision of previously generated actions. We evaluate the success rate of action code generation with respect to whether a player can interactively play through the entire generated story.

Problem

• Generating interactive fiction games using LLMs
• Dynamically creating actions for player freedom
• Ensuring game state consistency with generated actions

Innovation

• Generates story, world, and game code via LLM
• Uses LLM-generated preconditions for game state tracking
• Dynamically generates new actions for player creativity