Joseph Le Roux

Joseph Le Roux is an Associate Professor at the University of Paris 13, affiliated with the LIPN laboratory. His research focuses on Natural Language Processing (NLP), Machine Learning (ML), and their applications to optimization problems. He has supervised multiple PhD students, including Yash Kankanampati, Nicolas Floquet, and Francesco Demelas, among others.

Le Roux's work emphasizes dependency parsing, attention mechanisms, and optimization techniques in NLP. His research has been published in top venues such as NAACL, ICML, and EMNLP. He also contributes to academic service, including roles as an Action Editor for ARR and a member of the TAL journal's editorial board.

His teaching responsibilities include courses on neural networks, generative models in NLP, and programming for AI. He co-manages the Data Science hub at LIPN and leads research projects like SemiAmor (ANR CE23-2023-0005) and ParSiTi (ANR-16-CE33-0021).

Le Roux's academic contributions span over two decades, with a PhD in 2007 and an HDR (Habilitation) in 2024. His lab collaborations include work on XMG, a metagrammar compiler, and projects addressing challenges in social media NLP and MIP solving via ML.