Publications

Pregnancy AI: Development and Internal Validation of an Artificial Intelligence Tool to Predict Live Births in ICSI and IVF Cycles Using Clinical Features and Embryo Images

Jaume Minano Masip

Penelope Borduas

Isaac-Jacques Kadoch

Simon Phillips

Doina Precup

Daniel Dufort

2026-02-11

Medicina (published)

doi.org

Stabilizing Native Low-Rank LLM Pretraining

Paul Janson

Edouard Oyallon

Eugene Belilovsky

Foundation models have achieved remarkable success, yet their growing parameter counts pose significant computational and memory challenges.… (see more) Low-rank factorization offers a promising route to reduce training and inference costs, but the community lacks a stable recipe for training models from scratch using exclusively low-rank weights while matching the performance of the dense model. We demonstrate that Large Language Models (LLMs) can be trained from scratch using exclusively low-rank factorized weights for all non-embedding matrices without auxiliary"full-rank"guidance required by prior methods. While native low-rank training often suffers from instability and loss spikes, we identify uncontrolled growth in the spectral norm (largest singular value) of the weight matrix update as the dominant factor. To address this, we introduce Spectron: Spectral renormalization with orthogonalization, which dynamically bounds the resultant weight updates based on the current spectral norms of the factors. Our method enables stable, end-to-end factorized training with negligible overhead. Finally, we establish compute-optimal scaling laws for natively low-rank transformers, demonstrating predictable power-law behavior and improved inference efficiency relative to dense models.

2026-02-11

ArXiv (preprint)

arxiv.org

Affordances Enable Partial World Modeling with LLMs

Khimya Khetarpal

Gheorghe Comanici

Jonathan Richens

Jeremy Shar

Fei Xia

Laurent Orseau

Aleksandra Faust

Doina Precup

2026-02-10

ArXiv (preprint)

arxiv.org

Improving the Robustness of Large Language Models for Code Tasks via Fine-tuning with Perturbed Data

Yang Liu

Armstrong Foundjem

Xingfang Wu

Heng Li

Foutse Khomh

Context: In the fast-paced evolution of software development, Large Language Models (LLMs) have become indispensable tools for tasks such as… (see more) code generation, completion, analysis, and bug fixing. Ensuring the robustness of these models against potential vulnerabilities from handling diverse inputs is critical, as variations in input can lead to incorrect or insecure code outputs. Objective: This work aims to improve the robustness of LLMs for coding-related tasks against potential adversarial inputs. Specifically, we investigate how fine-tuning LLMs with perturbed datasets impacts their robustness against input perturbations. Method: We systematically evaluated LLM robustness by fine-tuning models using datasets perturbed at character-level, word-level, and sentence-level, comparing results against base models and models fine-tuned on unperturbed datasets. Results: Fine-tuning LLMs with perturbed datasets significantly improves model robustness (RD usually drops around 4\% - 6\%), especially for models with relatively weak robustness. However, this fine-tuning process typically results in a slight performance decrease (pass@1 usually drops around 1\% - 3\%) compared to fine-tuning with unperturbed datasets, although occasional performance improvements are observed. Conclusion \&Implications: Fine-tuning LLMs for coding tasks with perturbed data effectively enhances their robustness at the cost of a minor performance reduction, emphasizing the importance of balancing the robustness and performance of LLMs for coding applications.

2026-02-10

ArXiv (preprint)

arxiv.org

What Makes Value Learning Efficient in Residual Reinforcement Learning?

Guozheng Ma

Li Li

Haoyu Wang

Zixuan Liu

Pierre-Luc Bacon

Dacheng Tao

Residual reinforcement learning (RL) enables stable online refinement of expressive pretrained policies by freezing the base and learning on… (see more)ly bounded corrections. However, value learning in residual RL poses unique challenges that remain poorly understood. In this work, we identify two key bottlenecks: cold start pathology, where the critic lacks knowledge of the value landscape around the base policy, and structural scale mismatch, where the residual contribution is dwarfed by the base action. Through systematic investigation, we uncover the mechanisms underlying these bottlenecks, revealing that simple yet principled solutions suffice: base-policy transitions serve as an essential value anchor for implicit warmup, and critic normalization effectively restores representation sensitivity for discerning value differences. Based on these insights, we propose DAWN (Data-Anchored Warmup and Normalization), a minimal approach targeting efficient value learning in residual RL. By addressing these bottlenecks, DAWN demonstrates substantial efficiency gains across diverse benchmarks, policy architectures, and observation modalities.

2026-02-10

ArXiv (preprint)

arxiv.org

What do people want to fact-check?

Bijean Ghafouri

Dorsaf Sallami

Luca Luceri

Taylor Lynn Curtis

Jean-François Godbout

Emilio Ferrara

Reihaneh Rabbany

2026-02-10

ArXiv (preprint)

arxiv.org

Squeezing More from the Stream : Learning Representation Online for Streaming Reinforcement Learning

Nilaksh

Antoine Clavaud

Mathieu Reymond

Franccois Rivest

A. Chandar

AI Institute

Polytechnique Montr ´ eal

2026-02-09

ArXiv (preprint)

arxiv.org

The Untapped Potential of Food Webs in Systematic Conservation Planning

Louise M. J. O'Connor

Wilfried Thuiller

Ulrich Brose

Éléonore Chenevois

Carla Freund

Benoit Gauzens

Pierre Gaüzere

Catherine Graham

Michael Harfoot

Myriam R. Hirt

Sébastien Lavergne

Luigi Maiorano

Atte Moilanen

Peter H. Verburg

Piero Visconti

Laura J. Pollock

International conservation policy includes the dual aims of protecting biodiversity and nature's contributions to people (NCP). Achieving th… (see more)ese goals requires protecting not only species and habitats but also the networks of biotic interactions that sustain them. Food webs, which represent predator‐prey interactions between species, are increasingly recognised as a link between ecosystem structure, function, and resilience, which are concepts that are frequently cited in conservation policy. Yet, conservation planning and policy typically focus on individual species and habitats and overlook the interactions that support their persistence. We review the literature at the intersection of food web ecology and conservation, and highlight how food webs can inform three conservation goals: preventing species extinctions, maintaining ecosystem functions and NCP, and fostering ecosystem resilience. Food web data and metrics, such as interaction diversity, trophic diversity, connectance, or modularity, can be used to prioritize species that are key to ecosystem structure and functioning, and to guide spatial prioritization to protect functionally diverse and resilient communities. Given the growing availability of food web data, incorporating food webs in conservation planning can lead to more effective and resilient conservation outcomes that sustain biodiversity and ecosystem functions in the long term.

2026-02-09

Conservation Letters (published)

doi.org

Inverting Data Transformations via Diffusion Sampling

Jinwoo Kim

Sékou-Oumar Kaba

Jiyun Park

Seunghoon Hong

Siamak Ravanbakhsh

2026-02-08

ArXiv (preprint)

arxiv.org

HypRAG: Hyperbolic Dense Retrieval for Retrieval Augmented Generation

Hiren Madhu

Ngoc Bui

Ali Maatouk

Leandros Tassiulas

Smita Krishnaswamy

Menglin Yang 0001

Sukanta Ganguly

Kiran Srinivasan

Rex Ying

Embedding geometry plays a fundamental role in retrieval quality, yet dense retrievers for retrieval-augmented generation (RAG) remain large… (see more)ly confined to Euclidean space. However, natural language exhibits hierarchical structure from broad topics to specific entities that Euclidean embeddings fail to preserve, causing semantically distant documents to appear spuriously similar and increasing hallucination risk. To address these limitations, we introduce hyperbolic dense retrieval, developing two model variants in the Lorentz model of hyperbolic space: HyTE-FH, a fully hyperbolic transformer, and HyTE-H, a hybrid architecture projecting pre-trained Euclidean embeddings into hyperbolic space. To prevent representational collapse during sequence aggregation, we introduce the Outward Einstein Midpoint, a geometry-aware pooling operator that provably preserves hierarchical structure. On MTEB, HyTE-FH outperforms equivalent Euclidean baselines, while on RAGBench, HyTE-H achieves up to 29% gains over Euclidean baselines in context relevance and answer relevance using substantially smaller models than current state-of-the-art retrievers. Our analysis also reveals that hyperbolic representations encode document specificity through norm-based separation, with over 20% radial increase from general to specific concepts, a property absent in Euclidean embeddings, underscoring the critical role of geometric inductive bias in faithful RAG systems.

2026-02-07

ArXiv (preprint)

arxiv.org

BRIDGE: Predicting Human Task Completion Time From Model Performance

Mila - Québec

AI Institute

McGill University

Polytechnique Montréal

Periodic Labs

Servicenow Research

Canada Cifar

AI Chair

Evaluating the real-world capabilities of AI systems requires grounding benchmark performance in human-interpretable measures of task diffic… (see more)ulty. Existing approaches that rely on direct human task completion time annotations are costly, noisy, and difficult to scale across benchmarks. In this work, we propose BRIDGE, a unified psychometric framework that learns the latent difficulty scale from model responses and anchors it to human task completion time. Using a two-parameter logistic Item Response Theory model, we jointly estimate latent task difficulty and model capability from model performance data across multiple benchmarks. We demonstrate that latent task difficulty varies linearly with the logarithm of human completion time, allowing human task completion time to be inferred for new benchmarks from model performance alone. Leveraging this alignment, we forecast frontier model capabilities in terms of human task length and independently reproduce METR's exponential scaling results, with the 50% solvable task horizon doubling approximately every 6 months.

2026-02-05

ArXiv (preprint)

doi.org

arxiv.org

Constrained Group Relative Policy Optimization

Roger Girgis

Rodrigue De Schaetzen

Luke Rowe

Azal'ee Robitaille

Christopher Pal

Liam Paull

2026-02-04