Highlight Research
ICML 2026

42 Hi! PARIS Papers Accepted at ICML 2026

We are proud to highlight the strong presence of Hi! PARIS researchers at ICML 2026, one of the world’s leading conferences in machine learning.

This year, 42 papers involving Hi! PARIS affiliated researchers have been accepted, including oral, spotlight, poster, and regular contributions. This outstanding result reflects the depth, diversity, and scientific excellence of the Hi! PARIS research community.

The accepted works cover a wide range of topics, from optimization, diffusion models, flow matching, causal inference, bandits, graph learning, multimodal AI, and large language models to federated learning, uncertainty estimation, time series, robotics, and healthcare applications.

Beyond methodological advances, these contributions address concrete challenges in areas such as cybersecurity, finance, medical imaging, resource-efficient AI, trustworthy machine learning, and generative models. Many of them also reflect strong collaborations across the Hi! PARIS ecosystem, bringing together researchers from our institutions.

This collective achievement once again illustrates Hi! PARIS’s mission to advance AI research at the intersection of science, technology, business, and society.

Congratulations to our researchers!

Here is a list of papers accepted at ICML 2026 that include at least one author affiliated with Hi! PARIS:

Title Hi! PARIS Authors All Authors
Dynamic Programming for Epistemic Uncertainty in Markov Decision Processes Julien Grand-Clément, Alain Oliviero Durmus Axel Benyamine; Julien Grand-Clément; Marek Petrik; Michael I. Jordan; Alain Oliviero Durmus
Entropic Mirror Monte Carlo Alain Oliviero Durmus, Yohan Petetin Anas Cherradi; Yazid Janati; Alain Oliviero Durmus; Sylvain Le Corff; Yohan Petetin; Julien Stoehr
Feature-Aware (Hyper)graph Generation via Next-Scale Prediction Enzo Tartaglione, Jhony H. Giraldo Dorian Gailhard; Enzo Tartaglione; Lirida Naviner; Jhony H. Giraldo
GASS: Geometry-Aware Spherical Sampling for Disentangled Diversity Enhancement in Text-to-Image Generation Johannes F. Lutzeyer, Ye Zhu Ye Zhu; Kaleb Newman; Johannes F. Lutzeyer; Adriana Romero-Soriano; Michal Drozdzal; Olga Russakovsky
IdEst: Assessing Self-Supervised Learning Representations via Intrinsic Dimension Vicky Kalogeiton, Steve Oudot Julie Mordacq; Vicky Kalogeiton; Steve Oudot
MIRO: MultI-Reward cOnditioned pretraining improves T2I quality and efficiency Vicky Kalogeiton, David Picard Nicolas Dufour; Lucas Degeorge; Arijit Ghosh; Vicky Kalogeiton; David Picard
Multiple Choice Learning of Low-Rank Adapters for Language Modeling Mathieu Fontaine, Slim Essid, Gaël Richard Victor Letzelter; Hugo Malard; Mathieu Fontaine; Gaël Richard; Slim Essid; Andrei Bursuc; Patrick Perez
Spatiotemporal Imputation with Graph-Informed Flow Matching Aref Einizade, Jhony H. Giraldo Zepeng Zhang; Aref Einizade; Jhony H. Giraldo; Olga Fink
Tailoring Strictly Proper Scoring Rules for Downstream Tasks: An Application to Causal Inference Thomas Bonald, Matthieu Labeau, Gaël Varoquaux Roman Plaud; Alexandre Perez-Lebel; Antoine Saillenfest; Thomas Bonald; Marine Le Morvan; Gaël Varoquaux; Matthieu Labeau
TimeSAE: Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Models Quentin Bouniot, Stephan Clémençon Khalid Oublal; Quentin Bouniot; Qi Gan; Stephan Clémençon; Zeynep Akata
A Kinetic Energy Perspective of Flow Matching Michalis Vazirgiannis Ziyun Li; Huancheng Hu; Soon Hoe Lim; Xuyu Li; Fei Gao; Enmao Diao; Zezhen Ding; Michalis Vazirgiannis; Henrik Boström
A Tight Theory of Error Feedback Algorithms in Distributed Optimization Aymeric Dieuleveut Daniel Berg Thomsen; Adrien Taylor; Aymeric Dieuleveut
A Unifying View of Variational Generative Wasserstein Flows Anna Korba Paul Caucheteux; Clément Bonet; Anna Korba
Adaptive Bandit Algorithms for Contextual Matching Markets Vianney Perchet Shiyun Lin; Simon Mauras; Vianney Perchet; Nadav Merlis
Adaptive Momentum and Nonlinear Damping for Neural Network Training Gabriel Stoltz Aikaterini Karoni; Rajit Rajpal; Benedict J. Leimkuhler; Gabriel Stoltz
Amortized Maximum Inner Product Search with Learned Support Functions Marco Cuturi Theo X. Olausson; Joao Monteiro; Michal Klein; Marco Cuturi
Beyond ReLU: Bifurcation, Oversmoothing, and Topological Priors Maks Ovsjanikov Erkan Turan; Gaspard Abel; Maysam Behmanesh; Emery Pierson; Maks Ovsjanikov
Categorical Reparameterization with Denoising Diffusion Models Alain Oliviero Durmus Samson Gourevitch; Alain Oliviero Durmus; Eric Moulines; Jimmy Olsson; Yazid Janati
Diffusion Flow Matching: Dimension-Improved KL Bounds and Wasserstein Guarantees Alain Oliviero Durmus Marta Gentiloni Silveri; Giovanni Conforti; Alain Oliviero Durmus
Dimension-Free Multimodal Sampling via Preconditioned Annealed Langevin Dynamics Josselin Garnier Lorenzo Baldassari; Josselin Garnier; Knut Solna; Maarten V. de Hoop
Dissecting Multimodal In-Context Learning: Modality Asymmetries and Circuit Dynamics in modern Transformers Quentin Bouniot Yiran Huang; Karsten Roth; Quentin Bouniot; Wenjia Xu; Zeynep Akata
Graph Alignment via Dual-Pass Spectral Encoding and Latent Space Communication Maks Ovsjanikov Maysam Behmanesh; Erkan Turan; Maks Ovsjanikov
Joint Learning in the Gaussian Single Index Model Loucas Pillaud-Vivien Loucas Pillaud-Vivien; Adrien Schertzer
Learning High-Dimensional Parity Functions with Product Networks using Gradient Descent Hadi Ghauch Guillaume Larue; Louis-Adrien Dufrène; Quentin Lampin; Hadi Ghauch; Rekaya-Ben Othman
Learning Unmasking Policies for Diffusion Language Models Marco Cuturi Metod Jazbec; Theo X. Olausson; Louis Béthune; Pierre Ablin; Michael Kirchhof; Joao Monteiro; Victor Guilherme Turrisi da Costa; Jason Ramapuram; Marco Cuturi
Mantis: Lightweight Foundation Model for Time Series Classification Quentin Bouniot Vasilii Feofanov; Songkang Wen; Shifeng Xie; Simon Roschmann; Marius Alonso; Hongbo Guo; Romain Ilbert; Malik Tiomoko; Quentin Bouniot; Zeynep Akata; Lujia Pan; Jianfeng Zhang; Ievgen Redko
NetDiff: Graph Diffusion with Improved Global Capabilities to Generate and Update Mobile Network Topologies Cédric Adjih Félix Marcoccia; Victor Fagoo; Gilles Monzat de Saint Julien; Cédric Adjih; Thomas Watteyne; Paul Mühlethaler
Online Packet Scheduling with Deadlines and Learning Vianney Perchet Gianmarco Genalti; Achraf Azize; Vianney Perchet
Outcome-Aware Spectral Feature Learning for Instrumental Variable Regression Karim Lounici Dimitri Meunier; Jakub Wornbard; Vladimir R Kostic; Antoine Moulin; Alek Fröhlich; Karim Lounici; Massimiliano Pontil; Arthur Gretton
Random Process Flow Matching: Generative Implicit Representations of Multivariate Random Fields David Picard Julien Lalanne; David Picard; Lionel Boillot; Lina-María GUAYACÁN-CARRILLO; Leon Barens; Jean-Michel Pereira
Refined Analysis of Entropy-Regularized Actor-Critic Paul Mangold Safwan Labbi; Paul Mangold; Daniil Tiapkin; Eric Moulines
Removing Noise, not Finding Gold: Quality Filtering for Large-Scale Pretraining Marco Cuturi Thiziri Nait Saada; Louis Béthune; Michal Klein; David Grangier; Marco Cuturi; Pierre Ablin
Representation Learning for Equivariant Inference with Guarantees Karim Lounici Daniel Ordonez-Apraez; Vladimir R Kostic; Alek Fröhlich; Vivien Brandt; Karim Lounici; Massimiliano Pontil
Restoring Initial Noise Sensitivity in Text-to-Image Distillation through Geometric Alignment Ye Zhu Huayang Huang; Ruoyu Wang; Jinhui Zhao; Wei Deng; Daiguo Zhou; Jian Luan; Yu Wu; Ye Zhu
SOTAlign: Semi-Supervised Alignment of Unimodal Vision and Language Models via Optimal Transport Quentin Bouniot Simon Roschmann; Paul Krzakala; Sonia Mazelet; Quentin Bouniot; Zeynep Akata
The Art of Interrogation: Consistency Amplifies Factuality in Spatial Reasoning Maks Ovsjanikov Théo Uscidda; Marta Tintore Gazulla; Maks Ovsjanikov; Federico Tombari; Leonidas Guibas
The Latent Color Subspace: Emergent Order in High-Dimensional Chaos Quentin Bouniot Mateusz Pach; Jessica Bader; Quentin Bouniot; Serge Belongie; Zeynep Akata
Tightening the Score Matching Gap for Diffusion Models Alain Oliviero Durmus Benjamin Dupuis; Tyler Farghly; Maxime Haddouche; Alain Oliviero Durmus; Umut Simsekli
Toward Scalable and Valid Conditional Independence Testing with Spectral Representations Karim Lounici Alek Fröhlich; Vladimir R Kostic; Karim Lounici; Daniel Perazzo; Daniel Guimarães Tiezzi; Massimiliano Pontil
Unfolding Generative Flows with Koopman Operators: Trajectory-Preserving Linearization Maks Ovsjanikov Erkan Turan; Ari Siozopoulos; Louis Martinez; Julien Gaubil; Emery Pierson; Maks Ovsjanikov
Variance-Reduced $(\varepsilon, \delta)-$Unlearning using Forget Set Gradients El-Mahdi El-Mhamdi Martin Van Waerebeke; Giovanni Neglia; Kevin Scaman; Marco Lorenzi; El-Mahdi El-Mhamdi
Where Rectified Flows Leak: Characterising Membership Signals Along the Interpolation Path Geoffroy Peeters Thomas Sesmat; Gabriel Meseguer-Brocal; Geoffroy Peeters