ELLIS UnConference

A European Gathering on Machine Learning

December 8, 2026

Paris, France

About the event

The ELLIS UnConference is an annual community gathering, bringing together the European machine learning community to foster exchange, collaboration, and scientific discussion. Following several successful editions in Copenhagen, Spain, Germany and France, the event returns this year as a one-day gathering kicking off the NeurIPS satellite in Paris. Join us for a day of community building, scientific exchange, and the opportunity to present a paper recently accepted at a top machine learning venue.

Registration

Registration is now open!

Registration for the event and for the poster session takes place through a single form.

Register now

Accepted sessions

We are delighted to announce the 20 sessions selected for this year's program. Each session runs for two hours, in parallel with others, in the morning or afternoon slot.

Morning sessions

What Should a World Model Remember?

Randall Balestriero, Julia Kempe, Yann LeCun

  • World Models

Tabular Foundation Models: Principles, Practice, and Open Problems

Frank Hutter, Marine Le Morvan

  • Tabular Foundation Models

Multimodality for Everyone (M4E): Democratizing Multimodal Foundation Models

Sébastien Bratières, Yiannis Kompatsiaris, Luisa Bentivogli, et al.

  • Societal Aspects of AI

Health Foundation Models: From Prediction to Decision-Making and Self-Improvement

Jiancheng Yang, Kimmo Kartasalo, Azade Farshad, Koen Van Leemput

  • AI for Health

Are We at the Tipping Point for Agentic AI Deployment?

Nuria Oliver, Novi Quadrianto, Adrian Weller, Plamen P. Angelov, Javier Sanguino

  • Agentic AI

Long-Horizon AI Agents: Reasoning, Recovery, and Coordination

Beyza Ermis, Sharon Li

  • Agentic AI

Why Causality and Uncertainty Must Converge in Practice

Irene Balelli, Anna Calissano, Marianne Clausel, Emilie Devijver, Alex Markham, Adele Ribeiro

  • Causality

Open-Source Foundation Model Training in Europe

Aaron Klein, Michael Tiemann, Lapo Frati, Camille Denielle, Gopinath Balamurugan, Frank Hutter, David Salinas

  • Societal Aspects of AI

Constrained Learning for Scientific Modelling

Cecília Coelho, Andrzej Dulny, Fernanda Costa, Ricardo Baptista, Luís Ferrás, Oliver Niggemann

  • AI for Science

Afternoon sessions

Simulated vs. Unsimulated World Models

Sang Wan Lee, Mattia Rigotti, Seong-Whan Lee, Juan Helen Zhou, Jeehyun Kwag, Demian Battaglia

  • World Models

AI for Tabular & Relational Data: What’s Next?

Lennart Purucker, Madelon Hulsebos, David Holzmüller, Maximilian Schambach, Félix Lefebvre

  • Tabular Foundation Models
Rethinking AI: Sustainability and Frugality in the Era of Foundation Models

Quentin Bouniot, Alexandre Allauzen, Edouard Oyallon, Enzo Tartaglione, Florence d'Alché-Buc, Ivan Luiz de Moura Matos

  • Sustainable AI

AI for Earth and Climate Sciences

Gustau Camps-Valls, Markus Reichstein, Aishwarya Venkataramanan, Marieke Wesselkamp, Conrad Philipp

  • AI for Climate
Reflections on XAI

Nicolò Cesa-Bianchi, Emmanuel Esposito, Gunnar König, Ulrike von Luxburg, Maximilian Thiessen

  • Explainable AI

UniReps: Unifying Representations in Neural Models

Samuel Lippl, Clémentine Dominé, Zorah Lähner, Marco Fumero, Emily Cheng, Alexandra Proca, David Robin

  • Neural Representations

Amortized Probabilistic ML: How Neural Processes, Simulation-Based Inference and Tabular Foundation Models Are the Same Thing

Herilalaina Rakotoarison, Julia Linhart, Katharina Eggensperger, Luigi Acerbi

  • Probabilistic ML

ELLIS Innovation Area Kick-off: From Frontier Research to Impact

Matthias Bethge, Karteek Alahari, Pierre-Antoine Amiand-Leroy, Nuria Oliver, Fabian Theis, Tine Nørgaard Aage, Lisa Schöllhammer

  • Entrepreneurship in AI

Next Generation of Discovery Loops

Samuel Kaski, Chris Holmes, Yee Whye Teh, Haripriya Harikumar, Dmitry Kangin, Jaeyoung Lee, Ola Engkvist, Magnus Rattray, R de Santi

  • AI for Science
Learning on Graphs for Scientific Discovery

Daniel Daza, Victorita Dolean, Mykola Pechenizkiy

  • AI for Science

Call for posters

You can present a recently accepted paper (2026) at our poster session. We aim for top-quality papers in the field of ML (including but not restricted to: ICML, ICLR, AISTATS, JMLR). Come showcase your recent work in Paris!

Warning: you cannot present a paper accepted at NeurIPS 2026 at the UnConference.

Registration for the poster session happens together with the registration for the event.

Tentative schedule

Opening remarks

Keynote

Carina Prunkl

Coffee Break

Morning contributed sessions

Parallel sessions, see accepted sessions

Lunch Break

Afternoon contributed sessions

Parallel sessions, see accepted sessions

Coffee Break

Poster session

Contributed Posters

Sponsors

Interested in sponsoring the ELLIS UnConference? Get in touch at unconferencelocalchairs@gmail.com.

Tips for attendees

Conference venue

Palais des Congrès de Paris
2 Place de la Porte Maillot, 75017 Paris

Poster printing

Participants are responsible for printing their own posters.

Poster board specifications

  • Poster size: A1, portrait (59.4 × 84.1 cm).

Organisers

Organizing committee

Maxime Haddouche

CNRS

Marie Generali Lince

Inria, Crest, École polytechnique

Léo Dana

Inria, École normale supérieure

Matthieu Blanke

École normale supérieure

Sacha Braun

Inria, École normale supérieure

Eugène Berta

Inria, École normale supérieure

Ambroise Odonnat

Noah’s Ark Lab, Inria

Félix Chavelli

Inria, École normale supérieure

Advisory board

Edwige Cyffers

CNRS

Benjamin Guedj

University College London, Inria

Søren Hauberg

Technical University of Denmark

Conrad Philipp

Friedrich Schiller University of Jena