Gabriel Loiseau
Lille, France
Hi, I'm Gabriel.
I am a third year PhD student in Natural Language Processing in the Magnet team at Inria Lille.
My research focuses on practical applications of large language models for privacy.
My PhD is funded through France’s CIFRE program in partnership with Hornetsecurity, where I work as a Research Scientist and contribute to R&D initiatives involving LLMs.
I’m always happy to discuss NLP research, potential collaborations, and new ideas! Feel free to reach out to me by email or find me on X.
News
- June 2026. I'll be presenting our new ACL paper at the 16th Atelier sur la Protection de la Vie Privée in Trégastel (June 1-4).
- May 2026. I'll be at LREC 2026 to present our CALD-pseudo paper on distilling privacy sensitivity assessment from LLMs.
- April 2026. Our paper on adaptive text anonymization has been accepted to ACL 2026 (Findings).
- March 2026. Our paper on distilling privacy sensitivity assessment from LLMs has been accepted to CALD-pseudo (LREC 2026 workshop)! Check out the preprint.
- February 2026. New preprint on adaptive text anonymization: a novel task framework for context-aware, task-specific anonymization of text!
- January 2026. Our paper on Concept-Aligned Embeddings (CALE) has been accepted to EACL 2026 (oral).
- November 2025. I presented our work on Tau-Eval at EMNLP.
- September 2025. Our Tau-Eval paper has been accepted to EMNLP 2025.