EPFL Pre-ICML 2026 Poster Session

The ELLIS Lausanne Unit and the EPFL AI Center are pleased to host the EPFL Pre-ICML 2026 Poster Session, an informal research gathering highlighting recent contributions from the EPFL community to ICML 2026 and other leading machine learning conferences and workshops.

This poster session is a great opportunity for researchers to showcase their work, discover ongoing projects across EPFL’s machine learning ecosystem, and connect with fellow researchers ahead of ICML.

Registration required, present your poster: HERE (list of confirmed posters is provided below)

The session will primarily feature posters of papers accepted to ICML 2026. Priority will be given to:

1. ICML 2026 conference papers
2. ICML 2026 workshop papers
3. Other peer-reviewed papers from major machine learning conferences and workshops, subject to space availability

1. “A solvable high-dimensional model where nonlinear autoencoders learn structure invisible to PCA while test loss misaligns with generalization”, Vicente Conde Mendes∗ , Lorenzo Bardone∗ , Cedric Koller* , Jorge Medina Moreira∗ , Vittorio Erba, Emanuele Troiani, Lenka Zdeborova
2. “Single-Head Attention in High Dimensions: A Theory of Generalization, Weights Spectra, and Scaling Laws”, Fabrizio Boncoraglio, Vittorio Erba, Emanuele Troiani, Yizhou Xu, Lenka Zdeborová, Florent Krzakala
3. “Fixed-Point Masked Generative Modeling”, Andrea Miele, Yiming Qin, Alba Carballo-Castro, Justin Deschenaux, Pascal Frossard
4. “Weight Space Representation Learning via Neural Field Adaptation”, Zhuoqian Yang, Mathieu Salzmann, Sabine Süsstrunk          
5. “Preference-Based Distributed Welfare Maximization: A Game-Theoretic Approach”, Antoine Bergerault, Anna Maddux, Andreas Schlaginhaufen, Maryam Kamgarpour
6. “(1D) Ordered Tokens Enable Efficient Test-Time Search”, Zhitong Gao, Parham Rezaei, Ali Cy, Mingqiao Ye, Nataša Jovanović, Jesse Allardice, Afshin Dehghan, Amir Zamir, Roman Bachmann, Oğuzhan Fatih Kar                               
7. “RECTOR: Masked Region-Channel-Temporal Modeling for Affective and Cognitive Representation Learning”, Jinhan Liu, Mahsa Shoaran
8. "Sparse Neuron Ablation Triggers Catastrophic Collapse  of the Language Core in Vision-Language Models", Cen Lu, Yung-Chen Tang, Andrea Cavallaro 
9. "Recursive Scaling in Masked Diffusion Models Portrait", Alba Carballo-Castro, Julianna Piskorz, Paulius Rauba, Mihaela van der Schaar, Pascal Frossard
10. "Efficient Edge-aware Attention Network for Graph Generation", Vincent Jung, Alba Carballo-Castro, Yiming QIN, Lonneke van der Plas, Pascal Frossard
11. "PrivLEX: Detecting legal concepts in images through Vision-Language Models", Darya Baranouskaya, Andrea Cavallaro
12. "Why Linear Recurrent Memory Works in Partially Observable Reinforcement Learning", Yike Zhao, Onno Eberhard, Malek Khammassi, Ali H. Sayed, Michael Muehlebach
13. "Weight Space Representation Learning via Neural Field Adaptation", Zhuoqian Yang, Mathieu Salzmann, Sabine Süsstrunk
14. "Time series saliency maps: explaining models across multiple domains" - Christodoulos Kechris, Jonathan Dan, David Atienza
15. "Deep networks learn to parse uniform-depth context-free languages from local statistics" - Jack T. Parley, Francesco Cagnetta and Matthieu Wyart
16. "Multi Agent IL", Luca Viano, Freihaut, Nevali, Cevher, Geist and Ramponi
17. "DynaVieW: Schema-Guided World Modeling for Understanding Hierarchical Visual Dynamics", Silin Gao* , Hao Zhao*, Zeming Chen, Sepideh Mamooler, Antara Raaghavi Bhattacharya, Qiyu Wu, Hiromi Wakaki, Yuki Mitsufuji, Li Mi, Syrielle Montariol, Antoine Bosselut