I am a Ph.D. candidate at Linkmedia team of IRISA - CNRS, under the guidance of Guillaume Gravier and Pascale Sebillot. My research topic is about the explainability Deep Learning (or XAI - eXplainable AI) in the NLP (Natural Language Processing) domain application. Specifically, my current research investigate whether models using attention mechanism can provide human-like explanations (what we call the plausibility of explanations). Our intent is to use them to explain semantic links between texts in the Archival project¹.

My defense is expected in October 11, 2024, but I am currently available for a new job (permanent/CDI or post-doc). Ideally, I would like to work on topics closely related to my research fields (Natural Language Processing, Computer Vision and/or eXplainable AI), within an R&D department in the industry (I believe that science is more meaningful when used to solve real-world problems). However, I am open to all opportunities one may offer, as long as my contributions are to solve any problems.

Publications

Regularization, Semi-supervision, and Supervision for a Plausible Attention-Based Explanation

Duc Hau Nguyen, Cyrielle Mallart, Guillaume Gravier & Pascale Sébillot. Regularization, Semi-supervision, and Supervision for a Plausible Attention-Based Explanation. In Natural Language Processing and Information Systems (NLDB), pp. 285–298, 2023.

Géométrie de l’auto-attention en classification : Quand la géométrie remplace l’attention

Loïc Fosse, Duc Hau Nguyen, Pascale Sébillot, and Guillaume Gravier. Géométrie de l’auto-attention en classification : quand la géométrie remplace l’attention. In Conférence sur le Traitement Automatique des Langues Naturelles, pp. 137-15, 2023.

Hybridation Des Approches Symboliques et Apprentissage Profond Pour La Reconnaissance Des Entité Dans Les Signatures de Mail.

Duc Hau Nguyen, Nicolas Fouqué, Victor Klötzer, Hugo Thomas. Hybridation des approches symboliques et apprentissage profond pour la reconnaissance des entités dans les signatures de mail. In Atelier TextMine @ Extraction et Gestion des Connaissances, 2023

Filtrage et régularisation pour améliorer la plausibilité des poids d’attention dans la tâche d’inférence en langue naturelle

Duc Hau Nguyen, Guillaume Gravier, and Pascale Sébillot. Filtrage et régularisation pour améliorer la plausibilité des poids d’attention dans la tâche d’inférence en langue naturelle. Traitement Automatique des Langues Naturelles (TALN), pp. 95–103, 2022.

Géométrie de l’auto-Attention en classification : Quand la géométrie remplace l’attention

Loïc Fosse, Duc-Hau Nguyen, Pascale Sébillot, and Guillaume Gravier. Une étude statistique des plongements dans les modèles transformers pour le français. In Conference Traitement Automatique des Langues Naturelles (TALN), pp. 247–256, 2022.

Detecting Fake News Conspiracies with Multitask and Prompt-Based Learning

A Study of the Plausibility of Attention between RNN Encoders in Natural Language Inference

Duc Hau Nguyen, Guillaume Gravier, Pascale Sébillot. A Study of the Plausibility of Attention between RNN Encoders in Natural Language Inference. In IEEE International Conference on Machine Learning and Applications (ICMLA), pp. 1623–1629, 2021.

Professional experience

• 2020 – 2023: Doctoral researcher (3 years) at IRISA-CRNS (Rennes, France).
• 2018 – 2019: Industrial research apprenticeship (1 year) in Deep Learning at Michelin R&D center (Clermont-Ferrand, France).
• 2018 – 2019: Digital consultant (1 year) at Junior INSA Service (Toulouse, France).
• 2018: Research internship (4 months) at LAAS - CNRS (Toulouse, France).
• 2016: Development internship (5 months) at Bluesquare Computing (Paris, France).

Education

• 2024: Ph.D. in Computer Science at INSA Rennes (defense expected in October 2024), Rennes, France.
  • Title: Making AI Understandable for Humans: the Plausibility of Attention-Based Explanations in Natural Language Processing.
  • Major: Natural Language Processing, eXplainable Artificial Intelligence.
• 2019: M.Eng. in Computer Engineering at INSA Toulouse, Toulouse, France.
  • Major: Distributed systems and big data.
• 2017: Exchanged semester in Game Development at UQAC, Chicoutimi, Canada.
  • Courses: Game AI, Infographics, and Security.
• 2016: Associate degree (DUT) in Computer Science at IUT Paris Rives de Seine, Paris, France.
  • Ranking 2nd/150

Teaching