Pablo Piantanida

Biography

I am a professor at CentraleSupélec (Université Paris-Saclay) with the French National Centre for Scientific Research (CNRS), and Director of the International Laboratory on Learning Systems (ILLS) which gathers McGill University, École de technologie supérieure (ÉTS), Mila – Quebec AI Institute, France’s Centre Nationale de la Recherche Scientifique (CNRS), Université Paris-Saclay, and the École CentraleSupélec.

My research revolves around the application of advanced statistical and information-theoretic techniques to the field of machine learning. I am interested in developing rigorous techniques based on information measures and concepts for building safe and trustworthy AI systems and establishing confidence in their behavior and robustness, thereby securing their use in society. My primary areas of expertise include information theory, information geometry, learning theory, privacy, fairness, with applications to computer vision and natural language processing.

I obtained my undergraduate education at the University of Buenos Aires and pursued graduate studies in applied mathematics at Paris-Saclay University in France. Throughout my career, I have also held visiting positions at INRIA, Université de Montréal and Ecole de Technologie Supérieure (ÉTS), among others.

My earlier research encompassed the fields of information theory beyond distributed compression, statistical decision, universal source coding, cooperation, feedback, index coding, key generation, security, and privacy, among others.

I teach courses on machine learning, information theory and deep learning, covering topics such as statistical learning theory, information measures, statistical principles of neural networks.

Current Students

Chataigner Cléa

PhD - McGill University

Principal supervisor :

Golnoosh Farnadi

Maxime Darrin

PhD - McGill University

Principal supervisor :

PhD - École de technologie suprérieure

Matthieu Dubois

Collaborating researcher - Sorbonne université

PhD - École de technologie suprérieure

Fanny JOURDAN

Postdoctorate - École de technologie suprérieure

Co-supervisor :

Jackie Cheung

Website

Guillermo Martin Paula

PhD - École de technologie suprérieure

Ricard Marxer

Collaborating researcher - University of Toulon

Co-supervisor :

PhD - McGill University

Principal supervisor :

Mark Coates

Matteo Sammut

PhD - Université Paris Dauphine-PSL

Arthur SATOUF

Université Paris-Saclay

Amin Sghaier

Master's Research - École de technologie suprérieure

Co-supervisor :

Marco Pedersoli

Nicolas Thome

Collaborating researcher - Sorbonne Université

Website

Publications

Position: Auditing Is Not Evaluating; LLM Audit Requires Dynamic, Contextual, Budget-Aware and Reliable Evidence

Cléa Chataigner

Golnoosh Farnadi

Auditing large language models (LLMs) is increasingly urgent as these systems are deployed in high-stakes settings, yet existing evaluation … (see more)practices are ill-suited to meet auditing requirements. Directly repurposing standard evaluation tools can yield incomplete or misleading conclusions, e.g. overstating robustness when evidence comes from static prompts rather than adaptive, real-world interactions. This position paper argues that effective LLM audits must instead generate dynamic, context-sensitive, budget-aware, and reliable evidence. To support this position, we analyze how each of these principles can be operationalized through a four-component framework: Auditing Scope, Interactor, Evaluator, and Output. We highlight design requirements, assumptions, limitations and research directions, demonstrating how high-level principles can be translated into concrete, actionable, evidence-based procedures.

2026-02-05

2026 (preprint)

openalex.org

Happiness as a Measure of Fairness

Georg Pichler

Marco Romanelli

In this paper, we propose a novel fairness framework grounded in the concept of _happiness_, a measure of the utility each group gains from … (see more)decision outcomes. By capturing fairness through this intuitive lens, we not only offer a more human-centered approach, but also one that is mathematically rigorous: In order to compute the optimal, fair post-processing strategy, only a linear program needs to be solved. This makes our method both efficient and scalable with existing optimization tools. Furthermore, it unifies and extends several well-known fairness definitions, and our empirical results highlight its practical strengths across diverse scenarios.

2026-02-02

Artificial Intelligence and Statistics (poster)

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

Eric Aubinais

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)

Adapting Language Models to Produce Good Class Probabilities for Classification Tasks

Lautaro Estienne

Matias Vera

Elizabeth Fons

Elena Kochkina

LUCIANA FERRER

Large generative language models (GLM) provide a versatile tool for solving a wide variety of natural processing tasks. GLM responses, thoug… (see more)h, are provided in the form of text, without an indication of the model's confidence in the answer. This limits the usability of these models on high-risk applications where decisions made based on an incorrect answer can have severe consequences. In this work, we focus on the problem of generating class posterior distributions for text classification tasks like sentiment, news category and intent classification. These posteriors can be used for decision making and as interpretable scores for the user. We show that the naive approach for computing posteriors based on the token posteriors produced by the GLM results in extremely poor posteriors. We then explore different adaptation approaches for improving the quality of posteriors, focusing on low resource scenarios where a small amount of data is available for adaptation. We show that parameter-efficient supervised fine-tuning (SFT), while providing large gains in terms of decision quality, produces suboptimal posteriors due to overfitting. To address this problem, we propose an approach that combines SFT and post-hoc calibration (PHC) using a three-stage training strategy, improving the quality of both posteriors and categorical decisions.

2026-01-31

Transactions on Machine Learning Research (accepted)

BayesAdapter: enhanced uncertainty estimation in CLIP few-shot adaptation

Pablo Morales-Álvarez

Stergios Christodoulidis

Maria Vakalopoulou

Jose Dolz

The emergence of large pre-trained vision-language models (VLMs) represents a paradigm shift in machine learning, with unprecedented results… (see more) in a broad span of visual recognition tasks. CLIP, one of the most popular VLMs, has exhibited remarkable zero-shot and transfer learning capabilities in classification. To transfer CLIP to downstream tasks, adapters constitute a parameter-efficient approach that avoids backpropagation through the large model (unlike related prompt learning methods). However, CLIP adapters have been developed to target discriminative performance, and the quality of their uncertainty estimates has been overlooked. In this work we show that the discriminative performance of state-of-the-art CLIP adapters does not always correlate with their uncertainty estimation capabilities, which are essential for a safe deployment in real-world scenarios. We also demonstrate that one of such adapters is obtained through MAP inference from a more general probabilistic framework. Based on this observation we introduce BayesAdapter, which leverages Bayesian inference to estimate a full probability distribution instead of a single point, better capturing the variability inherent in the parameter space. In a comprehensive empirical evaluation we show that our approach obtains high quality uncertainty estimates in the predictions, standing out in calibration and selective classification. Our code will be publicly available upon acceptance of the paper.

2026-01-09

International Journal of Computer Vision (published)

Learning Task-Agnostic Representations through Multi-Teacher Distillation

Maxime Darrin

Banafsheh Karimian

Eric Granger

Jackie CK Cheung

Ismail Ben Ayed

Mohammadhadi Shateri

Casting complex inputs into tractable representations is a critical step across various fields. Diverse embedding models emerge from differe… (see more)nces in architectures, loss functions, input modalities and datasets, each capturing unique aspects of the input. Multi-teacher distillation leverages this diversity to enrich representations but often remains tailored to specific tasks. In this paper, we introduce a task-agnostic framework based on a ``majority vote" objective function. We demonstrate that this function is bounded by the mutual information between student and teachers' embeddings, leading to a task-agnostic distillation loss that eliminates dependence on task-specific labels or prior knowledge. Our evaluations across text, vision models, and molecular modeling show that our method effectively leverages teacher diversity, resulting in representations enabling better performance for a wide range of downstream tasks such as classification, clustering, or regression. Additionally, we train and release state-of-the-art embedding models, enhancing downstream performance in various modalities.

2025-09-17

NeurIPS.cc/2025/Conference (poster)

THUNDER: Tile-level Histopathology image UNDERstanding benchmark

Pierre Marza

Leo Fillioux

Sofiène Boutaj

KUNAL MAHATHA

Christian Desrosiers

Jose Dolz

Stergios Christodoulidis

Maria Vakalopoulou

Progress in a research field can be hard to assess, in particular when many concurrent methods are proposed in a short period of time. This … (see more)is the case in digital pathology, where many foundation models have been released recently to serve as feature extractors for tile-level images, being used in a variety of downstream tasks, both for tile- and slide-level problems. Benchmarking available methods then becomes paramount to get a clearer view of the research landscape. In particular, in critical domains such as healthcare, a benchmark should not only focus on evaluating downstream performance, but also provide insights about the main differences between methods, and importantly, further consider uncertainty and robustness to ensure a reliable usage of proposed models. For these reasons, we introduce THUNDER, a tile-level benchmark for digital pathology foundation models, allowing for efficient comparison of many models on diverse datasets with a series of downstream tasks, studying their feature spaces and assessing the robustness and uncertainty of predictions informed by their embeddings. THUNDER is a fast, easy-to-use, dynamic benchmark that can already support a large variety of state-of-the-art foundation, as well as local user-defined models for direct tile-based comparison. In this paper, we provide a comprehensive comparison of 23 foundation models on 16 different datasets covering diverse tasks, feature analysis, and robustness. The code for THUNDER is publicly available at https://github.com/MICS-Lab/thunder.

2025-09-17

NeurIPS.cc/2025/Datasets_and_Benchmarks_Track (spotlight)

Collaborative Rational Speech Act: Pragmatic Reasoning for Multi-Turn Dialog

Lautaro Estienne

Gabriel Ben Zenou

Nona Naderi

Jackie Chi Kit Cheung

As AI systems take on collaborative roles, they must reason about shared goals and beliefs-not just generate fluent language. The Rational S… (see more)peech Act (RSA) framework offers a principled approach to pragmatic reasoning, but existing extensions face challenges in scaling to multi-turn, collaborative scenarios. In this paper, we introduce Collaborative Rational Speech Act (CRSA), an information-theoretic (IT) extension of RSA that models multi-turn dialog by optimizing a gain function adapted from rate-distortion theory. This gain is an extension of the gain model that is maximized in the original RSA model but takes into account the scenario in which both agents in a conversation have private information and produce utterances conditioned on the dialog. We demonstrate the effectiveness of CRSA on referential games and template-based doctor-patient dialogs in the medical domain. Empirical results show that CRSA yields more consistent, interpretable, and collaborative behavior than existing baselines-paving the way for more pragmatic and socially aware language agents.

2025-07-17

arXiv (Cornell University) (preprint)

Rational Retrieval Acts: Leveraging Pragmatic Reasoning to Improve Sparse Retrieval

Arthur SATOUF

Gabriel Ben-Zenou

Benjamin Piwowarski

Habiboulaye Amadou-Boubacar

Current sparse neural information retrieval (IR) methods, and to a lesser extent more traditional models such as BM25, do not take into acco… (see more)unt the document collection and the complex interplay between different term weights when representing a single document. In this paper, we show how the Rational Speech Acts (RSA), a linguistics framework used to minimize the number of features to be communicated when identifying an object in a set, can be adapted to the IR case -- and in particular to the high number of potential features (here, tokens). RSA dynamically modulates token-document interactions by considering the influence of other documents in the dataset, better contrasting document representations. Experiments show that incorporating RSA consistently improves multiple sparse retrieval models and achieves state-of-the-art performance on out-of-domain datasets from the BEIR benchmark. https://github.com/arthur-75/Rational-Retrieval-Acts

2025-07-12

International ACM SIGIR Conference on Research and Development in Information Retrieval (published)

Multiple-model coding scheme for electrical signal compression

Corentin Presvôts

Michel Kieffer

Thibault Prevost

Patrick Panciatici

Zuxing Li

2025-03-31

Signal Processing (published)

A Strong Baseline for Molecular Few-Shot Learning

Hugo Jeannin

Ismail Ben Ayed

Few-shot learning has recently attracted significant interest in drug discovery, with a recent, fast-growing literature mostly involving con… (see more)voluted meta-learning strategies. We revisit the more straightforward fine-tuning approach for molecular data, and propose a regularized quadratic-probe loss based on the the Mahalanobis distance. We design a dedicated block-coordinate descent optimizer, which avoid the degenerate solutions of our loss. Interestingly, our simple fine-tuning approach achieves highly competitive performances in comparison to state-of-the-art methods, while being applicable to black-box settings and removing the need for specific episodic pre-training strategies. Furthermore, we introduce a new benchmark to assess the robustness of the competing methods to domain shifts. In this setting, our fine-tuning baseline obtains consistently better results than meta-learning methods.

2025-02-14

TMLR (accepted)

Membership Inference Risks in Quantized Models: A Theoretical and Empirical Study

Eric Aubinais

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 the original models. In this work, we investigate the impact of quantization procedures on the privacy of data-driven models, specifically focusing on their vulnerability to membership inference attacks. We derive an asymptotic theoretical analysis of Membership Inference Security (MIS), characterizing the privacy implications of quantized algorithm weights against the most powerful (and possibly unknown) attacks. Building on these theoretical insights, we propose a novel methodology to empirically assess and rank the privacy levels of various quantization procedures. Using synthetic datasets, we demonstrate the effectiveness of our approach in assessing the MIS of different quantizers. Furthermore, we explore the trade-off between privacy and performance using real-world data and models in the context of molecular modeling.

2025-02-09

ArXiv (preprint)