Publications

An Indicator of Membership Inference Security in Post-Training Quantized Models

Elisabeth Gassiat

Quantizing machine learning models has demonstrated its effectiveness in lowering memory and inference costs while maintaining performance l… (see more)evels comparable to those of the original models. In this work, we investigate the impact of quantization procedures on privacy in data-driven models, focusing on their vulnerability to membership inference attacks. Membership Inference Security (MIS) has recently been proposed to characterize the privacy of machine learning models against the most powerful (and possibly unknown) attacks. However, quantifying MIS appears to be computationally very difficult. In this paper, we propose a new MIS indicator for post-training quantization procedures of machine learning models that minimize an empirical loss. This new indicator is a byproduct of a theoretical asymptotic analysis of the MIS in this context. We also present a methodology for empirically estimating our MIS indicator. Using synthetic datasets and real-world data (in the context of drug discovery), we demonstrate the effectiveness of our approach in assessing and ranking the MIS of different quantizers.

2026-02-02

International Conference on Artificial Intelligence and Statistics (poster)

openreview.net

KQ-SVD: Compressing the KV Cache with Provable Guarantees on Attention Fidelity

Damien Lesens

Beheshteh T. Rakhshan

Guillaume Rabusseau

The Key–Value (KV) cache is central to the efficiency of transformer-based large language models (LLMs), storing previously computed vecto… (see more)rs to accelerate inference. Yet, as sequence length and batch size grow, the cache becomes a major memory bottleneck. Prior compression methods typically apply low-rank decomposition to keys alone or attempt to jointly embed queries and keys, but both approaches neglect that attention fundamentally depends on their inner products. In this work, we prove that such strategies are sub-optimal for approximating the attention matrix. We introduce KQ-SVD, a simple and computationally efficient method that directly performs an optimal low-rank decomposition of the attention matrix via a closed-form solution. By targeting the true source of redundancy, KQ-SVD preserves attention outputs with higher fidelity under compression. Extensive evaluations on LLaMA and Mistral models demonstrate that our approach consistently delivers superior projection quality.

2026-02-02

Artificial Intelligence and Statistics (poster)

doi.org

openreview.net

A Lightweight Library for Energy-Based Joint-Embedding Predictive Architectures

Basile Terver

Randall Balestriero

Megi Dervishi

David Fan

Quentin Garrido

Tushar Nagarajan

Koustuv Sinha

Wancong Zhang

Michael G. Rabbat

Yann Lecun

Amir Bar

We present EB-JEPA, an open-source library for learning representations and world models using Joint-Embedding Predictive Architectures (JEP… (see more)As). JEPAs learn to predict in representation space rather than pixel space, avoiding the pitfalls of generative modeling while capturing semantically meaningful features suitable for downstream tasks. Our library provides modular, self-contained implementations that illustrate how representation learning techniques developed for image-level self-supervised learning can transfer to video, where temporal dynamics add complexity, and ultimately to action-conditioned world models, where the model must additionally learn to predict the effects of control inputs. Each example is designed for single-GPU training within a few hours, making energy-based self-supervised learning accessible for research and education. We provide ablations of JEA components on CIFAR-10. Probing these representations yields 91% accuracy, indicating that the model learns useful features. Extending to video, we include a multi-step prediction example on Moving MNIST that demonstrates how the same principles scale to temporal modeling. Finally, we show how these representations can drive action-conditioned world models, achieving a 97% planning success rate on the Two Rooms navigation task. Comprehensive ablations reveal the critical importance of each regularization component for preventing representation collapse. Code is available at https://github.com/facebookresearch/eb_jepa.

2026-02-02

ArXiv (preprint)

arxiv.org

Lloyd's $K$-Means Clustering Algorithm is Frank-Wolfe in Disguise

Michael Pokojovy

J. Marcus Jobe

Simon Lacoste-Julien

Lloyd's …

2026-02-02

International Conference on Artificial Intelligence and Statistics (poster)

openreview.net

Local Inconsistency Resolution: The Interplay between Attention and Control in Probabilistic Models

Oliver Ethan Richardson

Joseph D Viviano

We present a generic algorithm for learning and approximate inference with an intuitive epistemic interpretation: iteratively focus on a sub… (see more)set of the model and resolve inconsistencies using the parameters under control. This framework, which we call Local Inconsistency Resolution (LIR) is built upon Probabilistic Dependency Graphs (PDGs), which provide a flexible representational foundation capable of capturing inconsistent beliefs. We show how LIR unifies and generalizes a wide variety of important algorithms in the literature, including the Expectation-Maximization (EM) algorithm, belief propagation, adversarial training, GANs, and GFlowNets. Each of these methods can be recovered as a specific instance of LIR by choosing a procedure to direct focus (attention and control). We implement this algorithm for discrete PDGs and study its properties on synthetically generated PDGs, comparing its behavior to the global optimization semantics of the full PDG.

2026-02-02

International Conference on Artificial Intelligence and Statistics (spotlight)

openreview.net

A multi-ancestry genetic reference for the Quebec population

Peyton McClelland

Georgette Femerling

Rose Laflamme

Alejandro Mejia-Garcia

Mohadese Sayahian Dehkordi

Hongyu Xiao

Alex Diaz-Papkovich

Justin Pelletier

Jean Christophe Grenier

Ken Sin Lo

Luke Anderson-Trocmé

Justin Bellavance

Vincent Chapdelaine

Genevieve Gagnon

Annelie De Mori

Gerardo Martinez

Kristen Mohler

Thibault de Malliard

Catherine Labbé

Marjorie Labrecque … (see 14 more)

Alexandre Montpetit

Dan Spiegelman

Guy A Rouleau

Jean-François Théroux

Hufeng Zhou

Simon L Girard

Julie G Hussin

Anne-Marie Laberge

Claude Bhérer

Martine Tétreault

Sarah A Gagliano Taliun

Daniel Taliun

Simon Gravel

Guillaume Lettre

While international efforts have characterized genetic variation in millions of individuals, the interplay of environmental, social, cultura… (see more)l, and genetic factors is poorly understood for most worldwide populations. The province of Quebec in Canada has been the site of numerous genetic studies, often focusing on individual Mendelian diseases in founder sub-populations. Here, we profiled and analyzed genome-wide genotyped variation in 29,337 Quebec residents from the large population-based cohort CARTaGENE (CaG), including rich phenotype and environmental data. We also sequenced the whole-genome of 2,173 CaG participants, including 163 and 132 individuals with grandparents born in Haiti and Morocco, respectively. We use this genetic information to gain insight into Quebec's demography and to help interpret the potential significance of variants identified in clinically important genes. We built an imputation panel by phasing the CaG whole-genome sequence data and showed, using genome-wide association studies (GWAS), how it improves the discovery of phenotype-genotype associations in this population. We provide allele frequency information and GWAS results through dedicated and publicly available websites. The genetic data, paired with phenotypic and environmental information, is also available for research use upon scientific and ethical review.

2026-02-02

Nature Communications (published)

doi.org

Observational Study of Maternal and Fetal Outcome in Posterior Reversible Encephalopathy Syndrome in Eclamptic Women in a Tertiary Care Institute

Prerna Kailashchand Gupta

Meenal Shailesh Sarmalkar

Madhuri A Mehendale

Aishwarya Agrawal

2026-02-02

Journal of South Asian Federation of Obstetrics and Gynaecology (published)

doi.org

Robust Intervention Learning from Emergency Stop Interventions

Ethan Pronovost

Khimya Khetarpal

Siddhartha Srinivasa

2026-02-02

ArXiv (preprint)

arxiv.org

The Role of Causal Features in Strategic Classification for Robustness and Alignment

Sophia Gunluk

Antonio Gois

Nir Rosenfeld

Nidhi Hegde

Simon Lacoste-Julien

Dhanya Sridhar

In strategic classification, an institution (e.g., a bank) anticipates adaptation from users who change their features to increase utility i… (see more)n a classification task (e.g., loan repayment). Since a key challenge is the distribution shift induced by users, we turn to causal models, which have been shown to bound the worst-case out-of-distribution (OOD) risk, and establish several new results that link causality and strategic classification. First, we show that causal classification leads to optimal classification error after any sufficiently large adaptation, when the noise is bounded in a certain way. Second, when these assumptions do not hold, we show OOD cross-entropy risk of optimal classifiers decomposes into an OOD bias term and a term arising from not using all observable features, allowing us to determine when causal classifiers have an advantage. Finally, we show that causal classifiers can align long-term incentives between institutions and users, contrasting with previous work that highlights social costs of such approaches. We validate our theory empirically on synthetic data, finding that our results predict behavior in practice.

2026-02-02

Artificial Intelligence and Statistics (poster)

openreview.net

On the Role of Depth in the Expressivity of RNNs

Maude Lizaire

Michael Rizvi-Martel

Éric Dupuis

Guillaume Rabusseau

The benefits of depth in feedforward neural networks (FNNs) are well known: composing multiple layers of linear transformations with nonline… (see more)ar activations enables complex computations. While similar effects are expected in recurrent neural networks (RNNs), it remains unclear how depth interacts with recurrence to shape expressive power. Here, we formally show that depth increases RNNs’ memory capacity efficiently with respect to parameters, enhancing expressivity both by enabling more complex input transformations and improving the retention of past information. We extend our analysis to 2RNNs, a generalization of RNNs with multiplicative interactions between inputs and hidden states. Unlike RNNs, which remain linear without nonlinear activations, 2RNNs perform polynomial transformations whose maximal degree grows with depth. We further show that multiplicative interactions cannot, in general, be replaced by layerwise nonlinearities. Finally, we validate these insights empirically on synthetic and real-world tasks.

2026-02-02

International Conference on Artificial Intelligence and Statistics (spotlight)

doi.org

openreview.net

Tractable Shapley Values and Interactions via Tensor Networks

Farzaneh Heidari

Chao Li

Guillaume Rabusseau

We show how to replace the …

2026-02-02

Artificial Intelligence and Statistics (poster)

doi.org