Publications

The list of publications from members of the team can be found below (*equal contribution).

Zero-shot Model-based Reinforcement Learning using Large Language Models Abdelhakim Benechehab, Youssef Attia El Hili, Ambroise Odonnat, Oussama Zekri, Albert Thomas, Giuseppe Paolo, Maurizio Filippone Ievgen Redko, Balázs Kégl ICLR 2025. arXiv / OpenReview / code /

Clustering Head: A Visual Case Study of the Training Dynamics in Transformers Ambroise Odonnat, Wassim Bouaziz, Vivien Cabannes Preprint. arXiv / code /

Easing Optimization Paths: A Circuit Perspective Ambroise Odonnat*, Wassim Bouaziz*, Vivien Cabannes ICASSP 2025. arXiv / code /

Large Language Models as Markov Chains Oussama Zekri*, Ambroise Odonnat*, Abdelhakim Benechehab, Linus Bleistein, Nicolas Boulle, Ievgen Redko Preprint. arXiv /

User-friendly Foundation Model Adapters for Multivariate Time Series Classification Romain Ilbert*, Vasilii Feofanov*, Malik Tiomoko, Themis Palpanas, Ievgen Redko Preprint. arXiv /

Analysing Multi-Task Regression via Random Matrix Theory with Application to Time Series Forecasting Romain Ilbert, Malik Tiomoko, Cosme Louart, Ambroise Odonnat, Vasilii Feofanov, Themis Palpanas, Ievgen Redko NeurIPS 2024 - Spotlight (top 10%). arXiv / OpenReview /

MaNo: Exploiting Matrix Norm for Unsupervised Accuracy Estimation Under Distribution Shifts Renchunzi Xie*, Ambroise Odonnat*, Vasilii Feofanov*, Weijian Deng, Jianfeng Zhang, Bo An NeurIPS 2024. arXiv / OpenReview / code /

SAMformer: Unlocking the Potential of Transformers in Time Series Forecasting with Sharpness-Aware Minimization and Channel-Wise Attention Romain Ilbert*, Ambroise Odonnat*, Vasilii Feofanov, Aladin Virmaux, Giuseppe Paolo, Themis Palpanas, Ievgen Redko ICML 2024 - Oral (top 5%). Also presented at CAp 2024 (oral). arXiv / OpenReview / code

Multi-class Probabilistic Bounds for Majority Vote Classifiers with Partially Labeled Data Vasilii Feofanov, Emilie Devijver, Massih-Reza Amini JMLR. Also presented at ICML 2024 (poster). Paper / Poster

Leveraging Ensemble Diversity for Robust Self-Training in the Presence of Sample Selection Bias Ambroise Odonnat, Vasilii Feofanov, Ievgen Redko AISTATS 2024. Also presented at CAp 2024 (long oral). arXiv / PMLR / code