Yannis Tevissen

Head of Science for Moments Lab. AI scientist.

yannis_tevissen.jpeg

I currently work as Head of Science for Moments Lab, where I lead Research initiatives on video understanding. More info can be found on the Moments Lab Research website.

Previously, I completed a PhD in artificial intelligence at the Institut Polytechnique de Paris, advised by Jérôme Boudy and Gérard Chollet.

My current research interests include video understanding, multimodal agents, and AI fairness.

Feel free to reach out to me to discuss any of these topics or to collaborate on research projects.

Recent news

Nov 06, 2025 I was featured in an Ad World News article and discussed my views about video understanding and AI-assisted content creation.
Nov 05, 2025 We presented a poster at BMVA Symposium on Multimodal Models about our latest paper on frame sampling for VLMS.
Jul 01, 2025 I’m joining the board of Universal Wings Mobility, an ambitious young company developing AVI, a conversational agent designed to help people with disabilities plan their travels.
Mar 02, 2025 I release UpToCure, a new platform for AI-generated reports on recent research towards curing rare diseases.
Dec 02, 2024 I was invited to speak at the OECD for a panel about AI and disability. I shared my concerns and recommendations for a more inclusive AI development.
Jul 10, 2024 I was awarded the Best Presentation Paper at the HSI 2024 conference for my talk about our recent paper: Towards Retrieval Augmented Generation over Large Video Libraries.
Jun 26, 2024 I was interviewed by TVBEurope about the future of video understanding together with Olivier Penin from TF1. Check out the interview.
Apr 01, 2024 We launched an AI Research Program with Moments Lab, focusing on video understanding.
Mar 28, 2024 I was invited to give a TEDx talk. I talked about disability biases in the society and how it propagates in AI systems. You can watch it here!
Dec 15, 2023 Successfully defended my PhD thesis on speaker diarization.

Science publications

  1. Video understanding
    Frame Sampling Strategies Matter: A Benchmark for small vision language models
    Marija Brkic, Anas Filali Razzouki, Yannis Tevissen, Khalil Guetari, and Mounim A El Yacoubi
    arXiv preprint arXiv:2509.14769, 2025
  2. Patent
    Computer-based platforms and methods for efficient AI-based digital video shot indexing
    Frédéric Petitpont, Philippe Petitpont, Yannis Tevissen, and Khalil Guetari
    Apr 2025
    US Patent 12,288,377
  3. Patent
    Systems and methods for AI generation of image captions enriched with multiple AI modalities
    Frédéric Petitpont, Yannis Tevissen, and Khalil Guetari
    Nov 2024
    US Patent 12,148,233
  4. Multimodal RAG
    Towards Retrieval Augmented Generation over Large Video Libraries
    Yannis Tevissen, Khalil Guetari, and Frédéric Petitpont
    In Proceedings of HSI 2024, Nov 2024
  5. AI Fairness
    Disability Representations: Finding Biases in Automatic Image Generation
    Yannis Tevissen
    In Workshop AVA: Accessibility, Vision, and Autonomy Meet, Nov 2024
  6. Video understanding
    Multimodal Chaptering for Long-Form TV Newscast Video
    Khalil Guetari, Yannis Tevissen, and Frédéric Petitpont
    Nov 2024
  7. VLM
    Inserting Faces inside Captions: Image Captioning with Attention Guided Merging
    Yannis Tevissen, Khalil Guetari, Marine Tassel, Erwan Kerleroux, and Frédéric Petitpont
    arXiv preprint arXiv:2405.02305, Nov 2024
  8. Edge AI
    Privacy Preserving Personal Assistant with On-Device Diarization and Spoken Dialogue System for Home and Beyond
    Gérard Chollet, Hugues Sansen, Yannis Tevissen, Jérôme Boudy, Mossaab Hariz, Christophe Lohr, and Fathy Yassa
    Proceedings of IHIET 2024, Nov 2024
  9. Speaker Diarization
    Diarisation multimodale: vers des modèles robustes et justes en contexte réel
    Yannis Tevissen
    Institut Polytechnique de Paris, Nov 2023
  10. Speaker Diarization
    Détection d’activité vocale Multi-flux pour la Diarisation du locuteur
    Yannis Tevissen, Jérôme Boudy, Gérard Chollet, and Frédéric Petitpont
    Proceedings of GRETSI 2023, Nov 2023
  11. Speaker Diarization
    Home monitoring for frailty detection through sound and speaker diarization analysis
    Yannis Tevissen, Dan Istrate, Vincent Zalc, Jérôme Boudy, Gérard Chollet, Frédéric Petitpont, and Sami Boutamine
    In JETSAN 2023, Nov 2023
  12. Fairness
    Towards measuring and scoring speaker diarization fairness
    Yannis Tevissen, Jérôme Boudy, Gérard Chollet, and Frédéric Petitpont
    arXiv preprint arXiv:2302.09991, Nov 2023
  13. Speech Processing
    Multi-stream voice activity detection for robust speaker diarization
    Yannis Tevissen, Jérôme Boudy, and Gérard Chollet
    In GDR ISIS 2022: Information, Signal, Image et ViSion: Traitement du signal pour la voix, Nov 2022
  14. Speech Processing
    The Newsbridge-Telecom SudParis VoxCeleb Speaker Recognition Challenge 2022 System Description
    Yannis Tevissen, Jérôme Boudy, and Frédéric Petitpont
    VoxCeleb Speaker Recognition Challenge 2022 Tack 4, Nov 2022