WorldGeneration

Generating Interactive Fiction worlds from story plots

From rajammanabrolu·Updated May 11, 2026·View on GitHub·

Code accompanying the paper ["Bringing Stories Alive: Generating Interactive Fiction Worlds"](http://arxiv.org/abs/2001.10161). The project is written primarily in Python, distributed under the MIT License license, first published in 2020. Key topics include: interactive-fiction, natural-language-generation, procedural-content-generation.

WorldGeneration

Code accompanying the paper "Bringing Stories Alive: Generating Interactive Fiction Worlds".

Neural PCG model (including AskBERT) is found in neural-based, Rule-based PCG model is found in rule-based, Evennia game generation framework is found in evennia-engine.

Each folder has its own README, follow the instructions in rule-based and neural-based to generate a *.dot file that can then be passed in the Evennia framework to create a playable game.

Dataset

The data used for finetuning fairytale and mystery models can be collected through /neural-based/scrape-wikipedia

BibTex

@article{ammanabrolu20world,
  title={Bringing Stories Alive: Generating Interactive Fiction Worlds},
  author={Ammanabrolu, Prithviraj and Cheung, Wesley and Tu, Dan and Broniec, William and Riedl, Mark O.},
  journal={CoRR},
  year={2020},
  url={http://arxiv.org/abs/2001.10161},
  volume={abs/2001.10161}
}