INFO : Initiation à Git
13 janvier 2026 @ 14 h 00 min – 16 h 00 min –
Campus Triolet, Pl. Eugène Bataillon, Bat 5, TD 5.209
Céline Mandier – Institut de Science des Données de Montpellier
Assistez à un atelier pratique pour maîtriser les bases de Git, un outil essentiel à la gestion et à la traçabilité des projets de recherche. Vous apprendrez à configurer Git, créer et gérer des dépôts, manipuler des fichiers et utiliser les commandes de base.
INFO, Git, Initiation
TabICL: A Tabular Foundation Model for In-Context Learning on Large Data
13 janvier 2026 @ 14 h 00 min –
Online
Machine Learning in Montpellier, Theory & Practice – Marine Le Morvan (Inria Saclay)
The long-standing dominance of gradient-boosted decision trees on tabular data is currently challenged by tabular foundation models using In-Context Learning (ICL): setting the training data as context for the test data and predicting in a single forward pass without parameter updates. While TabPFNv2 foundation model excels on tables with up to 10K samples, its alternating column- and row-wise attentions make handling large training sets computationally prohibitive. So, can ICL be effectively scaled and deliver a benefit for larger tables? We introduce TabICL, a tabular foundation model for classification, pretrained on synthetic datasets with up to 60K samples and capable of handling 500K samples on affordable resources. This is enabled by a novel two-stage architecture: a column-then-row attention mechanism to build fixed-dimensional embeddings of rows, followed by a transformer for efficient ICL. Across 200 classification datasets from the TALENT benchmark, TabICL is on par with TabPFNv2 while being systematically faster (up to 10 times), and significantly outperforms all other approaches. On 53 datasets with over 10K samples, TabICL surpasses both TabPFNv2 and CatBoost, demonstrating the potential of ICL for large data.
Machine Learning in Montpellier, Theory & Practice