SpinGPT: A Large-Language-Model Approach to Playing Poker Correctly
Advances in Computer Games 2025 This paper presents SpinGPT, a poker AI based on a large language model. Paper | Slides
Publications & talks
You can also find my articles on my Google Scholar profile.
Publications
Conference Papers
Preprints
Global PIQA: Evaluating Physical Commonsense Reasoning Across 100+ Languages and Cultures
Introducing Global PIQA, a new multilingual benchmark for 100+ languages. Paper | Code
Master's and Bachelor's theses
Detecting and quantifying uncertainty in the Studium Parisiense corpus
This Master’s thesis, written in French, studies how to detect and quantify the uncertainty expressed in the records of the Studium Parisiense prosopographical database, and converts them into a knowledge graph. Paper
Poker: the scientific journey of a controversial game, from the 18th century to today
This supervised research project, written in French, examines how a once disreputable card game became an object of scientific interest. Paper
Talks
SpinGPT - Training a Poker LLM
Talk, ACG'25, Online
At the Advances in Computer Games conference (21–23 October 2025), I presented the main results of my work on SpinGPT, a poker AI built on top of a large language model.
PokerGPT: Playing (Correct) Poker with a Large Language Model
Talk, PFIA'25, Dijon, France
I presented a preliminary version of my work on large language models for poker at PFIA’25, in the “Jeux et IA / MAFTEC” workshop in Dijon.