Published papers: 'Retrieval-Based Interleaved Visual Chain-of-Thought in Real-World Driving Scenarios' and 'Boosting Omnidirectional Stereo Matching with a Pre-trained Depth Foundation Model', with the former under review at NeurIPS 2025.
Research Experience
Before joining EPFL, worked as a Computer Vision Engineer at Preligens (now Safran.AI) and as a Research Intern at Heuritech. Led the creation of Helvipad, the first stereo depth estimation dataset based on 360° cameras.
Education
Ph.D. in Computer Science, March 2022, jointly from Conservatoire National des Arts et Métiers and valeo.ai, supervised by Prof. Nicolas Thome and Dr. Patrick Pérez. Master 2 Data Sciences from École Polytechnique in 2017, and a Master’s degree in computer science from Ecole Centrale de Lille in 2016.
Background
Senior Machine Learning Researcher at Raidium, focusing on advancing large multimodal models for medical applications. Previously, a postdoctoral researcher at EPFL in the Visual Intelligence for Transportation (VITA) lab, exploring how to leverage large multimodal models for autonomous driving.