Prompting for Performance: Exploring LLMs for Configuring Software
Helge Spieker, Théo Matricon, Nassim Belmecheri, Jørn Eirik Betten, Gauthier Le Bartz Lyan, Heraldo Borges, Quentin Mazouni, Dennis Gross, Arnaud Gotlieb, Mathieu Acher
Published in under review
EcoSearch: A Constant-Delay Best-First Search Algorithm for Program Synthesis
Théo Matricon, Nathanaël Fijalkow, Guillaume Lagarde
Published in AAAI 2025 A* (Oral, ~5% of submissions accepted for oral, 23% acceptance rate, 13k+ submissions)
Re-evaluating metamorphic testing of chess engines: A replication study
Axel Martin, Djamel Eddine Khelladi, Théo Matricon, Mathieu Acher
Published in IST
Theoretical foundations for programmatic reinforcement learning
Guruprerana Shabadi, Nathanaël Fijalkow, Théo Matricon
Published in GenPlan AAAI 2025 workshop
WikiCoder: Learning to Write Knowledge-Powered Code
Théo Matricon, Nathanaël Fijalkow, Gaëtan Margueritte
Published in SPIN (ETAPS)
Software: DeepSynth
Théo Matricon, Nathanaël Fijalkow, Guillaume Lagarde, Kevin Ellis
Published in JOSS
Challenges of Acquiring Compositional Inductive Biases via Meta-Learning
Marie Anastacio, Théo Matricon, Holger Hoos
Published in ECMLPKDD Workshop on Meta-Knowledge Transfer,
Scaling Neural Program Synthesis with Distribution-based Search
Nathanaël Fijalkow, Guillaume Lagarde, Théo Matricon, Kevin Ellis, Pierre Ohlmann, Akarsh Potta
Published in AAAI 2022 A* (Oral, ~5% of submissions accepted for oral, 15% acceptance rate, 9k+ submissions)