Uses WildChat
The Garden of Forking Prompts: How Users Explore Narrative Space in Story Generation
COLM 2026 · First arXiv submission
Introduces WildStories, containing 275,635 story prompts, and WildEdits, containing 24,291 reconstructed edit trees. The study follows revisions across separate conversations, organizes edits by direction and narrative target, and uses this framework to generate story-prompt variations for evaluation.
Data used
The yuntian-deng/WildChat-4.8M-Full release, filtered to 2,560,447 English prompt-response pairs with prompts of at least 20 words; one story-shape analysis uses WildChat-1M.
Findings
94.2% of prompts with an identified parent link to a different WildChat conversation, showing how story revisions span separate conversations.
Sections 3–4: sampling, inferred users, and edit-tree construction. Sections 5–6: revisions and prompt permutation. Read the source.
Dates and versions
First arXiv version. ArXiv version history.
COLM 2026 proceedings paper.
Cite this paper
Advait Deshmukh, Nora Benedict, Melanie Walsh, Maria Antoniak. 2026. The Garden of Forking Prompts: How Users Explore Narrative Space in Story Generation. Third Conference on Language Modeling.
Show BibTeX
@inproceedings{deshmukh2026forkingprompts,
title = {{The Garden of Forking Prompts: How Users Explore Narrative Space in Story Generation}},
author = {Deshmukh, Advait and Benedict, Nora and Walsh, Melanie and Antoniak, Maria},
year = {2026},
booktitle = {Third Conference on Language Modeling},
url = {https://openreview.net/forum?id=w6VGVCw4FF}
}