DraftFM: AI model predicts Magic: The Gathering picks before release

Researchers have developed DraftFM, a novel foundation model designed for the day-zero drafting of Magic: The Gathering expansions. This model can predict player card picks even before any actual gameplay data is available, relying solely on public card information. DraftFM achieved significant accuracy in predicting held-out picks and was even used to generate a card ranking for the unreleased set 'The Hobbit', which aligned well with expert reviews. AI

IMPACT This model's ability to predict outcomes with limited data could influence AI development in other domains requiring day-zero predictions.

RANK_REASON The cluster describes a research paper detailing a new foundation model for a specific application (game drafting). [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• DraftFM
• Gotit.pub
• Hugging Face
• IArxiv
• Magic: The Gathering
• Magic: The Gathering Arena
• ScienceCast
• The Hobbit

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →