Publications

FLAM: Frame-Wise Language-Audio Modeling
Yusong Wu
Christos Tsirigotis
Ke Chen
Cheng-Zhi Anna Huang
Aaron Courville
Oriol Nieto
Prem Seetharaman
Justin Salamon
Recent multi-modal audio-language models (ALMs) excel at text-audio retrieval but struggle with frame-wise audio understanding. Prior works … (voir plus)use temporal-aware labels or unsupervised training to improve frame-wise capabilities, but they still lack fine-grained labeling capability to pinpoint when an event occurs. While traditional sound event detection models can precisely localize events, they are limited to pre-defined categories, making them ineffective for real-world scenarios with out-of-distribution events. In this work, we introduce FLAM, an open-vocabulary contrastive audio-language model capable of localizing specific sound events. FLAM employs a memory-efficient and calibrated frame-wise objective with logit adjustment to address spurious correlations, such as event dependencies and label imbalances during training. To enable frame-wise supervision, we leverage a large-scale dataset with diverse audio events, LLM-generated captions and simulation. Experimental results and case studies demonstrate that FLAM significantly improves the open-vocabulary localization capability while maintaining strong performance in global retrieval and downstream tasks.
2025-07-14
International Conference on Machine Learning (Accept (poster))
doi.org
proceedings.mlr.press
Neural activity resolved in space and time through fusion of large-scale EEG and fMRI datasets.
Ian Charest
Peter Brotherwood
Mathias Salvas-Hébert
Kendrick Kay
Frédéric Gosselin
2025-07-14
Journal of Vision (publié)
doi.org
Personalizing brain stimulation: continual learning for sleep spindle detection
Milo Sobral
Hugo R Jourde
S Ehsan M Bajestani
Emily B J Coffey
Giovanni Beltrame
Objective. Personalized stimulation, in which algorithms used to detect neural events adapt to a user’s unique neural characteristics, may… (voir plus) be crucial to enable optimized and consistent stimulation quality for both fundamental research and clinical applications. Precise stimulation of sleep spindles-transient patterns of brain activity that occur during non rapid eye movement sleep that are involved in memory consolidation-presents an exciting frontier for studying memory functions; however, this endeavour is challenged by the spindles’ fleeting nature, inter-individual variability, and the necessity of real-time detection. Approach. We tackle these challenges using a novel continual learning framework. Using a pre-trained model capable of both online classification of sleep stages and spindle detection, we implement an algorithm that refines spindle detection, tailoring it to the individual throughout one or more nights without manual intervention. Main results. Our methodology achieves accurate, subject-specific targeting of sleep spindles and enables advanced closed-loop stimulation studies. While fine-tuning alone offers minimal benefits for single nights, our approach combining weight averaging demonstrates significant improvement over multiple nights, effectively mitigating catastrophic forgetting. Significance. This work represents an important step towards signal-level personalization of brain stimulation that can be applied to different brain stimulation paradigms including closed-loop brain stimulation, and to different neural events. Applications in fundamental neuroscience may enhance the investigative potential of brain stimulation to understand cognitive processes such as the role of sleep spindles in memory consolidation, and may lead to novel therapeutic applications.
2025-07-14
Journal of Neural Engineering (publié)
doi.org
Relative Entropy Pathwise Policy Optimization
Claas Voelcker
Axel Brunnbauer
Marcel Hussing
Michal Nauman
Pieter Abbeel
Eric R. Eaton
Radu Grosu
Amir-massoud Farahmand
Igor Gilitschenski
Score-function policy gradients have delivered strong results in game-playing, robotics and language-model fine-tuning. Yet its high-varianc… (voir plus)e often undermines training stability. On the other hand, pathwise policy gradients alleviate the training variance, but are reliable only when driven by an accurate action-conditioned value function which is notoriously hard to train without relying on past off-policy data. In this paper, we discuss how to construct a value-gradient driven, on-policy algorithm that allow training Q-value models purely from on-policy data, unlocking the possibility of using pathwise policy updates in the context of on-policy learning. We show how to balance stochastic policies for exploration with constrained policy updates for stable training, and evaluate important architectural components that facilitate accurate value function learning. Building on these insights, we propose Relative Entropy Pathwise Policy Optimization (REPPO), an efficient on-policy algorithm that combines the sample-efficiency of pathwise policy gradients with the simplicity and minimal memory footprint of standard on-policy learning. We demonstrate that REPPO provides strong empirical performance at decreased sample requirements, wall-clock time, memory footprint as well as high hyperparameter robustness in a set of experiments on two standard GPU-parallelized benchmarks.
2025-07-14
ArXiv (prépublication)
doi.org
arxiv.org
Scaling Trends in Language Model Robustness
Nikolaus Howe
Ian McKenzie
Oskar Hollinsworth
Michał Zając
Tom Tseng
Aaron Tucker
Pierre-Luc Bacon
Adam Gleave
Increasing model size has unlocked a dazzling array of capabilities in modern language models. At the same time, even frontier models remain… (voir plus) vulnerable to jailbreaks and prompt injections, despite concerted efforts to make them robust. As both attack and defense gain access to more compute, and as models become larger, what happens to robustness? We argue that to answer this question requires a \emph{scaling} approach, which we employ in an extensive study of language model robustness across several classification tasks, model families, and adversarial attacks. We find that in the absence of explicit safety training, larger models are not consistently more robust; however, scale improves sample efficiency in adversarial training, though it worsens compute efficiency. Further, we find that increasing attack compute smoothly improves attack success rate against both undefended and adversarially trained models. Finally, after exploring robustness transfer across attacks and threat models, we combine attack and defense scaling rates to study the offense-defense balance. We find that while attack scaling outpaces adversarial training across all models studied, larger adversarially trained models might give defense the advantage in the long run. These results underscore the utility of the scaling lens, and provide a paradigm for evaluating future attacks and defenses on frontier models.
2025-07-14
International Conference on Machine Learning (Accept (spotlight poster))
doi.org
proceedings.mlr.press
STAGED: A Multi-Agent Neural Network for Learning Cellular Interaction Dynamics
João Felipe Rocha
Ke Xu
Xingzhi Sun
Ananya Krishna
Dhananjay Bhaskar
Blanche Mongeon
Morgan Craig
Mark B. Gerstein
Smita Krishnaswamy
The advent of single-cell technology has significantly improved our understanding of cellular states and subpopulations in various tissues u… (voir plus)nder normal and diseased conditions by employing data-driven approaches such as clustering and trajectory inference. However, these methods consider cells as independent data points of population distributions. With spatial transcriptomics, we can represent cellular organization, along with dynamic cell-cell interactions that lead to changes in cell state. Still, key computational advances are necessary to enable the data-driven learning of such complex interactive cellular dynamics. While agent-based modeling (ABM) provides a powerful framework, traditional approaches rely on handcrafted rules derived from domain knowledge rather than data-driven approaches. To address this, we introduce Spatio Temporal Agent-Based Graph Evolution Dynamics(STAGED) integrating ABM with deep learning to model intercellular communication, and its effect on the intracellular gene regulatory network. Using graph ODE networks (GDEs) with shared weights per cell type, our approach represents genes as vertices and interactions as directed edges, dynamically learning their strengths through a designed attention mechanism. Trained to match continuous trajectories of simulated as well as inferred trajectories from spatial transcriptomics data, the model captures both intercellular and intracellular interactions, enabling a more adaptive and accurate representation of cellular dynamics.
2025-07-14
ArXiv (prépublication)
doi.org
arxiv.org
The Courage to Stop: Overcoming Sunk Cost Fallacy in Deep Reinforcement Learning
Jiashun Liu
Johan Obando-Ceron
Pablo Samuel Castro
Aaron Courville
Ling Pan
Off-policy deep reinforcement learning (RL) typically leverages replay buffers for reusing past experiences during learning. This can help i… (voir plus)mprove sample efficiency when the collected data is informative and aligned with the learning objectives; when that is not the case, it can have the effect of "polluting" the replay buffer with data which can exacerbate optimization challenges in addition to wasting environment interactions due to wasteful sampling. We argue that sampling these uninformative and wasteful transitions can be avoided by addressing the sunk cost fallacy, which, in the context of deep RL, is the tendency towards continuing an episode until termination. To address this, we propose learn to stop (LEAST), a lightweight mechanism that enables strategic early episode termination based on Q-value and gradient statistics, which helps agents recognize when to terminate unproductive episodes early. We demonstrate that our method improves learning efficiency on a variety of RL algorithms, evaluated on both the MuJoCo and DeepMind Control Suite benchmarks.
2025-07-14
International Conference on Machine Learning (Accept (poster))
doi.org
proceedings.mlr.press
The Impact of On-Policy Parallelized Data Collection on Deep Reinforcement Learning Networks
Walter Mayor
Johan Obando-Ceron
Aaron Courville
Pablo Samuel Castro
The use of parallel actors for data collection has been an effective technique used in reinforcement learning (RL) algorithms. The manner in… (voir plus) which data is collected in these algorithms, controlled via the number of parallel environments and the rollout length, induces a form of bias-variance trade-off; the number of training passes over the collected data, on the other hand, must strike a balance between sample efficiency and overfitting. We conduct an empirical analysis of these trade-offs on PPO, one of the most popular RL algorithms that uses parallel actors, and establish connections to network plasticity and, more generally, optimization stability. We examine its impact on network architectures, as well as the hyper-parameter sensitivity when scaling data. Our analyses indicate that larger dataset sizes can increase final performance across a variety of settings, and that scaling parallel environments is more effective than increasing rollout lengths. These findings highlight the critical role of data collection strategies in improving agent performance.
2025-07-14
International Conference on Machine Learning (Accept (poster))
doi.org
proceedings.mlr.press
UI-Vision: A Desktop-centric GUI Benchmark for Visual Perception and Interaction
Shravan Nayak
Xiangru Jian
Kevin Qinghong Lin
Juan A. Rodriguez
Montek Kalsi
Rabiul Awal
Nicolas Chapados
M. Tamer Özsu
Aishwarya Agrawal
David Vázquez
Christopher Pal
Perouz Taslakian
Spandana Gella
Sai Rajeswar
Human Annotator
Autonomous agents that navigate Graphical User Interfaces (GUIs) to automate tasks like document editing and file management can greatly enh… (voir plus)ance computer workflows. While existing research focuses on online settings, desktop environments, critical for many professional and everyday tasks, remain underexplored due to data collection challenges and licensing issues. We introduce UI-Vision, the first comprehensive, license-permissive benchmark for offline, fine-grained evaluation of computer use agents in real-world desktop environments. Unlike online benchmarks, UI-Vision provides: (i) dense, high-quality annotations of human demonstrations, including bounding boxes, UI labels, and action trajectories (clicks, drags, and keyboard inputs) across 83 software applications, and (ii) three fine-to-coarse grained tasks—Element Grounding, Layout Grounding, and Action Prediction—with well-defined metrics to rigorously evaluate agents’ performance in desktop environments. Our evaluation reveals critical limitations in state-of-the-art models like UI-TARS-72B, including issues with understanding professional software, spatial reasoning, and complex actions like drag-and-drop. These findings highlight the challenges in developing fully autonomous computer-use agents. With UI-Vision, we aim to advance the development of more capable agents for real-world desktop tasks.
2025-07-14
International Conference on Machine Learning (Accept (poster))
doi.org
proceedings.mlr.press
Effects of a Virtual Reality Hypnosis Intervention on Chronic Pain: A User Experience and Proof-of-concept Study
Alexandra Chevestrier-Lefeuvre
Joséphine Guiné
Jade Véronneau
Julie Lebeau
Floriane Rousseaux
Audrey Laurin
Marie-Fania Simard
Nadia Godin
Karim Jerbi
Philippe Richebé
Mathieu Landry
Pierre Rainville
Valentyn Fournier
David Ogez
Abstract

Chronic pain is a significant public health issue in Canada, with approximately one in four Canadians ove… (voir plus)r the age of 15 living with this condition. Due to its impact on individuals—both physically and psychologically—and its financial burden on the healthcare system, it is crucial to develop cost-effective and efficient treatment methods. Hypnosis and virtual reality have emerged as promising solutions in this context. This study aims to evaluate the preliminary efficacy and feasibility of an intervention combining virtual reality and hypnosis. The study involved 30 patients with chronic pain who were invited to test a hypnosis application delivered through a virtual reality device. Levels of pain, anxiety, and relaxation were measured before and after the intervention, while satisfaction, cybersickness, and user experience were evaluated post-intervention. At the end of the intervention, participants were invited to participate in a semi-structured interview to provide feedback on their satisfaction with the experience. Participants reported high levels of satisfaction with the intervention, a positive user experience, and minimal symptoms of cybersickness. The intervention was effective in reducing anxiety (W = 173.5, p = .002) and pain (W = 253.5, p< .001) while significantly enhancing relaxation levels (W = 9.00, p< .001). This intervention demonstrated effectiveness in reducing pain and anxiety while improving relaxation levels among individuals with chronic pain, paving the way for further investigations of the involved mechanisms.

2025-07-13
Research Square (prépublication)
doi.org
Frequency enrichment of coding variants in a French-Canadian founder population and its implication for inflammatory bowel diseases
Claude Bhérer
Jean-Christophe Grenier
Justin Pelletier
Gabrielle Boucher
Genevieve Gagnon
Philippe Goyette
Dariel Ashton-Beaucage
Christine Stevens
Robert Battat
Alain Bitton
Philippe M Campeau
Catherine Laprise
Quebec IBD Genetics Consortium
Hailiang Huang
Mark Daly
Daniel Taliun
Julie G Hussin
Vincent Mooser
John D Rioux
The genetic features of founder populations with recent bottlenecks, causing some deleterious variants to rise to higher frequencies, can en… (voir plus)hance the power of rare variant association studies. French Canadians from Quebec represent a recent founder population with a particular disease heritage comprising more than 30 prevalent Mendelian conditions. Here, we characterize coding variation in this founder population using exome sequencing data from 2,820 French-Canadian participants - patients with inflammatory bowel diseases (IBD), parents and controls from the Quebec IBD cohort. We find that 18% of rare coding variants are 10-100 times more frequent than in non-Finnish Europeans (NFE). A total of 4,133 missense and loss-of-function variants were significantly enriched with a median 28-fold enrichment, revealing the potential for genotype-phenotype associations in this population. We describe significantly enriched pathogenic variants, including those known to account for the increased prevalence of rare diseases in FC compared to other European descent populations, such as Agenesis of corpus callosum and peripheral neuropathy (SLC12A6) and Leigh Syndrome French Canadian type (LRPPRC). Finally, we investigate whether rare protein-coding variants, enriched in French Canadians by the founder effect, contribute to the risk of IBD using trio and case/control cohorts. In addition to replicating associations in NOD2 and IL23R, we identified new candidate association signals, including enriched variants in SLC35E3, and ARSA. Our findings show that, even in well-characterized founder populations like the French Canadians, there remains untapped potential for genetic discovery, revealing both rare and complex disease risk factors through enriched coding variation.
2025-07-13
medRxiv (prépublication)
doi.org
Integrating equity, diversity, and inclusion throughout the lifecycle of artificial intelligence for healthcare: a scoping review
Ting Wang
Elham Emami
Dana Jafarpour
Raymond Tolentino
Genevieve Gore
S. A. Rahimi
The lack of Equity, Diversity, and Inclusion (EDI) principles in the lifecycle of Artificial Intelligence (AI) technologies in healthcare is… (voir plus) a growing concern. Despite its importance, there is still a gap in understanding the initiatives undertaken to address this issue. This review aims to explore what and how EDI principles have been integrated into the design, development, and implementation of AI studies in healthcare. We followed the scoping review framework by Levac et al. and the Joanna Briggs Institute. A comprehensive search was conducted until April 29, 2022, across MEDLINE, Embase, PsycInfo, Scopus, and SCI-EXPANDED. Only research studies in which the integration of EDI in AI was the primary focus were included. Non-research articles were excluded. Two independent reviewers screened the abstracts and full texts, resolving disagreements by consensus or by consulting a third reviewer. To synthesize the findings, we conducted a thematic analysis and used a narrative description. We adhered to the PRISMA-ScR checklist for reporting scoping reviews. The search yielded 10,664 records, with 42 studies included. Most studies were conducted on the American population. Previous research has shown that AI models improve when socio-demographic factors such as gender and race are considered. Despite frameworks for EDI integration, no comprehensive approach systematically applies EDI principles in AI model development. Additionally, the integration of EDI into the AI implementation phase remains under-explored, and the representation of EDI within AI teams has been overlooked. This review reports on what and how EDI principles have been integrated into the design, development, and implementation of AI technologies in healthcare. We used a thorough search strategy and rigorous methodology, though we acknowledge limitations such as language and publication bias. A comprehensive framework is needed to ensure that EDI principles are considered throughout the AI lifecycle. Future research could focus on strategies to reduce algorithmic bias, assess the long-term impact of EDI integration, and explore policy implications to ensure that AI technologies are ethical, responsible, and beneficial for all.
2025-07-13
PLOS Digital Health (publié)
doi.org