October 10, 2025 : Théo Gigant has brilliantly defended his PhD thesis on automatic summarization of audiovisual documents... Congrats !!
June 30, 2025: Théo Gigant, Élise Lincker, Théodora Pazakou, and Elisabeth Olamisan will be attending TALN 2025 to present our papers "Mitigating the Impact of Reference Quality on the Evaluation of Summarization Systems through Reference-Free Metrics" and "Visual Accessibility and Inclusive Education: A Preliminary Study on Alt-Text Generation."
June 11, 2025: Ana Rodriguez will present our work on bias detection in gender identification models for Japanese data at the "Demographic Diversity in Computer Vision" workshop at CVPR.
October 10, 2024: The papers "Towards Inclusive Education: Multimodal Classification of Textbook Images for Accessibility" co-authored with Saumya Yadav, "AD2AT: Audio Description to Alternative Text, a Dataset of Alternative Text from Movies" co-authored with Élise Lincker and "Evaluating VQA Models' Consistency in the Scientific Domain" co-authored with Khanh-An Quan and Pr. Shin'ichi Satoh will be presented at the MMM2025 conference in Nara in January 2025
September 20, 2024: The paper Mitigating the Impact of Reference Quality on Evaluation of Summarization Systems with Reference-Free Metrics, co-authored with Théo Gigant, Frédéric Dufaux, and Marc Decombas, has been accepted to EMNLP 2024
July 30, 2024: The MINERAL project on multi-dimensional modeling of conversations, in collaboration with Orange Innovation and LIS, has received funding from the ANR