Accueil

Inspirer le développement de l'intelligence artificielle au bénéfice de tous·tes

Un professeur s'entretient avec ses étudiants dans un café/lounge.

Situé au cœur de l’écosystème québécois en intelligence artificielle (IA), Mila rassemble une communauté de plus de 1400 personnes spécialisées en apprentissage automatique et dédiées à l’excellence scientifique et l’innovation.

À propos

À la une

Nouvelles

22 Oct 2026
L’IA au service de gestion des feux de forêt au Canada
Vue d'une forêt à vol d'oiseau
Lire la nouvelle
08 Oct 2026
Mila et Inria renforcent la coopération en recherche sur l’IA entre le Canada et la France
Bruno Sportisse, président-directeur général d’Inria, et Valérie Pisano, présidente et cheffe de la direction de Mila
Lire la nouvelle
06 Oct 2026
Cinq membres de Mila nommé·e·s chaires en IA Canada-CIFAR
Photo des 5 membres Mila titulaires d'une chaire en IA Canada-CIFAR
Lire la nouvelle
Voir plus de nouvelles

Corps professoral

Fondé en 1993 par le professeur Yoshua Bengio, Mila regroupe aujourd'hui plus de 140 professeur·e·s affilié·e·s à l'Université de Montréal, l'Université McGill, Polytechnique Montréal et HEC Montréal. L'institut accueille également des professeur·e·s de l'Université Laval, de l'Université de Sherbrooke, de l'École de technologie supérieure (ÉTS) et de l'Université Concordia.

Consultez l'annuaire en ligne

Photo de Yoshua Bengio

Publications récentes

Adaptive Order Policies for Masked Diffusion
Jama Hussein Mohamud
Mohsin Hasan
Mirco Ravanelli
Yoshua Bengio
Masked diffusion models have seen great success in capturing data distributions over discrete sequences in domains such as text and proteins… (voir plus). These models generate data by iteratively unmasking tokens starting from a fully masked sequence, with the unmasking order typically chosen at random or using a heuristic based on denoiser probabilities. In this work, we propose a scheme for learning the unmasking order using an additional lightweight policy network on top of a diffusion model. Our proposed loss reweights terms in the masked diffusion loss according to policy probabilities, and results in a policy that prefers positions where the denoiser is more likely to be correct. We study this loss in two settings: (i) training solely the policy while using a frozen pre-trained denoiser, and (ii) training the policy and denoiser jointly with the weighted loss to allow for mutual adaptation. We demonstrate that our approach outperforms common heuristics on problems that are sensitive to token ordering, such as combinatorial problems, proteins as well as various coding and language tasks.
2026-12-30
OpenReview (publié)
doi.org
openreview.net
Building2Building: A Large Scale Benchmark for Generalizable Real-World Reinforcement Learning
Vincent Taboga
Justin Veilleux
Doseok Jang
Anushree Rankawat
Pierre-Luc Bacon
Reinforcement learning (RL) has achieved strong results in control, yet learned policies remain brittle to changes in dynamics, action space… (voir plus)s, observation spaces, or goals, a critical limitation for real-world deployment. Existing benchmarks offer limited diversity and complexity, making it difficult to rigorously study transfer, multi-task learning, and meta-learning in RL. We introduce Building2Building (B2B), a large-scale suite of realistic Heating, Ventilation, and Air Conditioning (HVAC) control environments built on EnergyPlus, a state-of-the-art building simulator. B2B is fully compatible with the Gymnasium interface and features a parametric building generator, enabling the systematic generation of diverse building configurations with heterogeneous observation and action spaces. Based on this suite, we define benchmark tasks targeting key open challenges in RL, including goal adaptation, dynamics adaptation, action-space shifts, and cross-domain transfer. By providing a large-scale, diverse, and physically grounded testbed with standardized evaluation protocols, B2B enables systematic investigation of generalization and transfer in continuous control. Beyond advancing research on generalization in RL, this new benchmark also carries significant societal implications by enabling improved HVAC control at scale, one of the most energy-intensive systems in buildings.
2026-12-30
OpenReview (publié)
doi.org
openreview.net
DASB - Discrete Audio and Speech Benchmark
Pooneh Mousavi
Jarod Duret
Darius Petermann
Artem Ploujnikov
Luca Della Libera
Anastasia Kuznetsova
Cem Subakan
Mirco Ravanelli
Duret, Jarod
Kuznetsova, Anastasia
Discrete audio tokens have recently gained considerable attention for their potential to bridge audio and language processing, enabling mult… (voir plus)imodal language models that can both generate and understand audio. However, preserving key information such as phonetic content, speaker identity, and paralinguistic cues remains a major challenge. Identifying the optimal tokenizer and configuration is further complicated by inconsistent evaluation settings across existing studies. To address this, we introduce the Discrete Audio and Speech Benchmark (DASB), a comprehensive framework for benchmarking discrete audio tokens across speech, general audio, and music domains on a range of discriminative and generative tasks. Our results show that discrete representations are less robust than continuous ones and require careful tuning of factors such as model architecture, data size, learning rate, and capacity. Semantic tokens generally outperform acoustic tokens, but a gap remains between discrete tokens and continuous features, highlighting the need for further research. DASB codes, evaluation setup, and leaderboards are publicly available at https://poonehmousavi.github.io/DASB-website/.
2026-12-30
OpenReview (publié)
doi.org
openreview.net
Do Tabular Foundation Models Know Physics? Contamination, Units, and the Deterministic Limit
Wassim Tenachi
Yashar Hezaveh
Laurence Perreault Levasseur
Pierre-Luc Bacon
Tabular foundation models (TFMs) learn to fill in tables the way language models fill in text, and tables are arguably the format in which m… (voir plus)ost physical measurement arrives. Did they learn any physics in the process? They are Bayesian by construction, so the question is what their prior contains. We probe it directly, evaluating four of them (TabPFN-3, TabICLv2, TabDPT and Real-TabPFN-2.5) against six baselines on datasets sampled from 316 physical equations, in and out of domain. TFMs dominate, out of the box and after tuning. But we show that their prior can represent neither a noiseless mechanism nor physical units, which is why they interpolate physics without yet being able to act as physical models.
2026-12-30
OpenReview (publié)
doi.org
openreview.net
Voir plus de publications
Mila Ventures

Mila Ventures

Notre branche de capital de risque cultive la prochaine génération d'entreprises, soutenues par l'écosystème de recherche en IA de classe mondiale de Mila. Nous investissons dans des fondateurs·rices visionnaires qui bâtissent à la frontière des technologies de pointe, de l'IA, des STIM et au-delà.

Nous croyons que l'avenir sera façonné par les entrepreneur·e·s scientifiques.

En savoir plus

Une personne regarde un ciel étoilé.