We are delighted to announce that our paper has been accepted at ECML PKDD 2026 (Applied Data Science Track), a KIISE CS Top Conference!

Modeling Matches as Language: A Generative Transformer Approach for Counterfactual Player Valuation in Football Miru Hong, Minho Lee, Geonhee Jo, Hyeokje Cho, Hyunsung Kim, Pascal Bauer, and Sang-Ki Ko

The paper introduces ScoutGPT, a generative model that treats football match events as sequential tokens within a language modeling framework. Built on a NanoGPT-based Transformer trained on next-token prediction, ScoutGPT simulates event sequences under hypothetical lineups and uses Monte Carlo sampling to enable counterfactual transfer simulation, capturing player-specific impact beyond traditional static metrics.

Huge congratulations to Miru Hong for leading this work, and to all the co-authors on this great achievement! 🎉