Publications

REVE: A Foundation Model for EEG -- Adapting to Any Setup with Large-Scale Pretraining on 25,000 Subjects

Yassine El Ouahidi

Jonathan Lys

Philipp Thölke

Nicolas Farrugia

Bastien Pasdeloup

Vincent Gripon

Karim Jerbi CoCo Lab

Giulia Lioi

2025-10-23

ArXiv (prépublication)

doi.org

arxiv.org

Surrogate-based quantification of policy uncertainty in generative flow networks

Ram'on Nartallo-Kaluarachchi

Robert Manson-Sawko

Shashanka Ubaru

Dongsung Huh

Malgorzata J. Zimo'n

Lior Horesh

Yoshua Bengio

2025-10-23

ArXiv (prépublication)

doi.org

arxiv.org

Generative Point Tracking with Flow Matching

Mattie Tesfaldet

Adam W. Harley

Konstantinos G. Derpanis

D. Nowrouzezahrai

Christopher Pal

Tracking a point through a video can be a challenging task due to uncertainty arising from visual obfuscations, such as appearance changes a… (voir plus)nd occlusions. Although current state-of-the-art discriminative models excel in regressing long-term point trajectory estimates -- even through occlusions -- they are limited to regressing to a mean (or mode) in the presence of uncertainty, and fail to capture multi-modality. To overcome this limitation, we introduce Generative Point Tracker (GenPT), a generative framework for modelling multi-modal trajectories. GenPT is trained with a novel flow matching formulation that combines the iterative refinement of discriminative trackers, a window-dependent prior for cross-window consistency, and a variance schedule tuned specifically for point coordinates. We show how our model's generative capabilities can be leveraged to improve point trajectory estimates by utilizing a best-first search strategy on generated samples during inference, guided by the model's own confidence of its predictions. Empirically, we evaluate GenPT against the current state of the art on the standard PointOdyssey, Dynamic Replica, and TAP-Vid benchmarks. Further, we introduce a TAP-Vid variant with additional occlusions to assess occluded point tracking performance and highlight our model's ability to capture multi-modality. GenPT is capable of capturing the multi-modality in point trajectories, which translates to state-of-the-art tracking accuracy on occluded points, while maintaining competitive tracking accuracy on visible points compared to extant discriminative point trackers.

2025-10-22

ArXiv (prépublication)

doi.org

arxiv.org

High IL1R1 expression predicts poor survival and benefit from stem cell transplant in intermediate-risk acute myeloid leukemia from the Leucegene cohort

Guillaume Richard-Carpentier

Francois Béliveau

Sandrine Lacoste

Banafsheh Khakipoor

Véronique Lisi

Michael Vladovsky

Miriam Marquis

Jean-François Spinella

Patrick Gendron

Sébastien Lemieux

Vincent-Philippe Lavallee

Guy Sauvageau

Josée Hébert

There is an unmet clinical need to identify patients with acute myeloid leukemia and intermediate-risk cytogenetics who benefit from allogen… (voir plus)eic hematopoietic stem cell transplantation in first remission, especially among those without FLT3 -ITD mutation. We analyzed transcriptomic data from the Leucegene cohort composed of 316 patients with acute myeloid leukemia and intermediate-risk cytogenetics who have been treated with intensive chemotherapy. We evaluated associations between gene expression and overall survival or relapse-free survival and we analyzed the interaction between gene expression and allogeneic hematopoietic stem cell transplantation to identify biomarkers that predict the benefit of stem cell transplantation in this subgroup of patients. We identified high IL1R1 expression ( IL1R1 high ) as a prognostic and predictive marker in the Leucegene cohort. IL1R1 high (≥ 2.0 transcripts per million) was associated with older age, monocytic differentiation, higher frequency of FLT3 -ITD and RUNX1 mutations and lower frequency of IDH1 / 2 and bZIP CEBPA mutations. Patients with IL1R1 high had lower 5-year overall survival (10% vs 38%, p < 0.01), and higher 5-year cumulative incidence of relapse (76% vs 59%, p < 0.01) than those with low IL1R1 expression. IL1R1 high was independently associated with overall survival in multivariable analyses including age, white blood cell count at diagnosis and NPM1 , FLT3 -ITD, bZIP CEBPA , RUNX1 , ASXL1 and DNMT3A mutations (HR 1.78, p < 0.01). Importantly, in landmark analysis, hematopoietic stem cell transplantation in first remission significantly improved 5-year overall survival in patients with IL1R1 high (67% vs 27%, HR 0.33, p < 0.01), but not in patients with IL1R1 low (62% vs 54%, HR 0.72, p = 0.31), especially among those without FLT3 -ITD mutation (48% vs 50%, HR 0.93, p = 0.85). In patients who proceeded to allogeneic hematopoietic stem cell transplantation, the 5-year overall survival was 60% in patients with IL1R1 high compared to 56% in patients with IL1R1 low confirming that the worse prognosis associated with high expression of IL1R1 was abrogated by stem cell transplantation. IL1R1 expression is a candidate marker to identify patients with intermediate-risk cytogenetics acute myeloid leukemia at high risk of relapse who benefit from allogeneic hematopoietic stem cell transplantation in first remission.

2025-10-22

Biomarker Research (publié)

doi.org

The Intermodal Railroad Blocking and Railcar Fleet-Management Planning Problem

Julie Kienzle

Serge Bisaillon

T. Crainic

Emma Frejinger

Rail is a cost-effective and relatively low-emission mode for transporting intermodal containers over long distances. This paper addresses t… (voir plus)actical planning of intermodal railroad operations by introducing a new problem that simultaneously considers three consolidation processes and the management of a heterogeneous railcar fleet. We model the problem with a scheduled service network design with resource management (SSND-RM) formulation, expressed as an integer linear program. While such formulations are challenging to solve at scale, we demonstrate that our problem can be tackled with a general-purpose solver when provided with high-quality warm-start solutions. To this end, we design a construction heuristic inspired by a relax-and-fix procedure. We evaluate the methodology on realistic, large-scale instances from our industrial partner, the Canadian National Railway Company: a North American Class I railroad. The computational experiments show that the proposed approach efficiently solves practically relevant instances, and that solutions to the SSND-RM formulation yield substantially more accurate capacity estimations compared to those obtained from simpler baseline models. Managerial insights from our study highlight that ignoring railcar fleet management or container loading constraints can lead to a severe underestimation of required capacity, which may result in costly operational inefficiencies. Furthermore, our results show that the use of multi-platform railcars improves overall capacity utilization and benefits the network, even if they can locally lead to less efficient loading as measured by terminal-level slot utilization performance indicators.

2025-10-22

ArXiv (prépublication)

arxiv.org

BugPilot: Complex Bug Generation for Efficient Learning of SWE Skills

Atharv Sonwane

Isadora White

Hyunji Lee

Matheus Pereira

Lucas Caccia

Minseon Kim

Zhengyan Shi

Chinmay Singh

Alessandro Sordoni

Marc-Alexandre Côté

Xingdi Yuan

2025-10-21

ArXiv (prépublication)

doi.org

arxiv.org

Controllable Generation of Drug-like Molecules with Multi-modal Variational Flow

Fang Sun

Zhihao Zhan

Hongyu Guo

Ming Zhang

Jian Tang

Yizhou Sun

Designing drug molecules that bind effectively to target proteins while maintaining desired pharmacological properties remains a fundamental… (voir plus) challenge in drug discovery. Current approaches struggle to simultaneously control molecular topology and 3D geometry, often requiring expensive retraining for new design objectives. We propose a multi-modal variational flow framework that addresses these limitations by integrating a 2D topology encoder with a 3D geometry generator. Our architecture encodes molecular graphs into a learned latent distribution via junction tree representations, then employs normalizing flows to autoregressively generate atoms in 3D space conditioned on the protein binding site. This design enables zero-shot controllability: by manipulating the latent prior distribution, we can generate molecules with specific substructures or optimized properties without model retraining. Experiments on the CrossDocked benchmark show that our model achieves 31.1% high-affinity rate, substantially outperforming existing methods, while maintaining superior drug-likeness and structural diversity. Our framework opens new possibilities for on-demand molecular design, allowing medicinal chemists to rapidly explore chemical space with precise control over both structural motifs and physicochemical properties.

2025-10-21

logconference.io/LOG/2025/Conference (poster)

openreview.net

A Derandomization Framework for Structure Discovery: Applications in Neural Networks and Beyond

Nikos Tsikouras

Yorgos Pantis

Ioannis Mitliagkas

Christos Tzamos

2025-10-21

ArXiv (prépublication)

doi.org

arxiv.org

Improving the Physics of Video Generation with VJEPA-2 Reward Signal

Jianhao Yuan

Xiaofeng Zhang

Felix Friedrich

Nicolas Beltran-Velez

Melissa Hall

Reyhane Askari Hemmat

Xiaochuang Han

Nicolas Ballas

Michal Drozdzal

Adriana Romero

2025-10-21

ArXiv (prépublication)

doi.org

arxiv.org

International AI Safety Report: First Key Update, Capabilities and Risk Implications

Prof. Yoshua Bengio

Stephen Clare

Carina Prunkl

Maksym Andriushchenko

BEN BUCKNALL

Philip Fox

Tiancheng Hu

Cameron Jones

Sam Manning

Nestor Maslej

Vasilios Mavroudis

Conor McGlynn

Malcolm Murray

Shalaleh Rismani

Charlotte Stix

Lucia Velasco

Nicole Wheeler

Daniel Privitera

Sören Mindermann

Daron Acemoglu … (voir 36 de plus)

Thomas G. Dietterich

Fredrik Heintz

Geoffrey Hinton

Nick Jennings

Susan Leavy

Teresa Ludermir

Vidushi Marda

Helen Margetts

John McDermid

Jane Munga

Arvind Narayanan

Alondra Nelson

Clara Neppel

Sarvapali D. (Gopal) Ramchurn

Stuart Russell

Marietje Schaake

Bernhard Schölkopf

Alvaro Soto

Lee Tiedrich

Gael Varoquaux

Andrew Yao

Ya-Qin Zhang

Lambrini Das

Claire Dennis

Arianna Dini

Freya Hempleman

Samuel Kenny

Patrick King

Hannah Merchant

Jamie-Day Rawal

Rose Woolhouse

The field of AI is moving too quickly for a single yearly publication to keep pace. Significant changes can occur on a timescale of months, … (voir plus)sometimes weeks. This is why we are releasing Key Updates: shorter, focused reports that highlight the most important developments between full editions of the International AI Safety Report. With these updates, we aim to provide policymakers, researchers, and the public with up-to-date information to support wise decisions about AI governance. This first Key Update focuses on areas where especially significant changes have occurred since January 2025: advances in general-purpose AI systems' capabilities, and the implications for several critical risks. New training techniques have enabled AI systems to reason step-by-step and operate autonomously for longer periods, allowing them to tackle more kinds of work. However, these same advances create new challenges across biological risks, cyber security, and oversight of AI systems themselves. The International AI Safety Report is intended to help readers assess, anticipate, and manage risks from general-purpose AI systems. These Key Updates ensure that critical developments receive timely attention as the field rapidly evolves.

2025-10-21

SuperIntelligence - Robotics - Safety & Alignment (publié)

doi.org

Random Forest Autoencoders for Guided Representation Learning

Kevin R. Moon

Jake S. Rhodes

Extensive research has produced robust methods for unsupervised data visualization. Yet supervised visualization…

2025-10-21