Summary

AI Researcher at Illuin Technology (Paris). MSc in Artificial Intelligence from the University of Surrey. Research interests include multimodal AI, NLP, retrieval-augmented generation, and speech processing.

Education

• Artificial Intelligence

  2024
  University of Surrey

  Courses: Deep Learning, Computer Vision, Natural Language Processing, Machine Learning
• General Engineering

  2024
  Ecole Centrale de Lyon

  Courses: Artificial Intelligence, Machine Learning, Probability and Statistics, Chaos and Fractals
• Mathematics, Physics, Computer Science (MPSI-MP)

  2021
  College Stanislas Paris VI

  Courses: Mathematics, Physics, Computer Science, Chemistry, Engineering Sciences

Work Experience

• Experienced Data Scientist

  2025-09-01 -
  Illuin Technology

  Development of artificial intelligence solutions for industrial clients.

  • NLP research (visual embeddings for industrial contexts)
  • Speech processing research (model evaluation, speech-to-text model finetuning)
  • Software development of cutting-edge AI solutions
• Data Scientist & Python Developer Intern

  2023-01-01 - 2023-12-31
  AI-vidence

  Research and development internship in AI explainability.

  • Development of a Python library for ML/AI model explainability
  • Documentation writing, creation of explanatory videos and designs
  • Participation in SpringTech Paris-Saclay 2023 at HEC

Skills

Publications

• ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios

  2026
  arXiv preprint (arXiv:2601.08620)

  A comprehensive multimodal RAG benchmark featuring multi-type queries over visually rich document corpora. Covers 10 datasets across diverse professional domains, comprising ~26,000 document pages paired with 3,099 human-verified queries in 6 languages.

  View Publication
• Working Notes on Late Interaction Dynamics: Analyzing Targeted Behaviors of Late Interaction Models

  2026
  First Late Interaction Workshop (LIR) @ ECIR 2026

  An analysis of targeted behaviors of late interaction models, presented at the First Late Interaction Workshop (LIR) at ECIR 2026.

  View Publication
• Exploring Zero-Shot Capabilities of Multi-Modal Foundation Models for Semantic Image Segmentation

  2024
  Master's Thesis, University of Surrey

  An exploration of zero-shot capabilities of multi-modal foundation models applied to semantic image segmentation tasks.

  View Publication

Languages

• French

  Native
• English

  Fluent (C1 — TOEFL ITP 623/677)
• Spanish

  Intermediate (B1)

Interests

• AI Explainability
• Computer Vision and Sports
• Multimodal AI
• Causality