Amphithéâtre Maurice Halbwachs, Site Marcelin Berthelot
Open to all
-

Abstract

The development of architectures exploiting " deep " neural learning methods in machine translation has led to a considerable increase in the acceptability and usability of machine-computed translations. These new architectures have also made it possible to implement machine translation devices that go beyond the usual framework of translating a source-language text into a target-language text : direct translation of speech, joint translation of text and image, and so on. In this talk, I will present one such device, designed to translate from multiple source languages into multiple source languages, highlighting the computational and linguistic benefits that such multilingual translation systems bring, particularly for translating from and into minority languages.

François Yvon

François Yvon

François Yvon is Director of Research at CNRS, and has been with the MLIA team at the Institut des Systèmes Intelligents et de Robotique (ISIR/CNRS and Sorbonne Université since July 2023. D. in Computer Science from ENST (1996), he was recruited as a lecturer in the Computer Science and Networks department, before being appointed Professor of Computer Science at Université Paris-Sud in 2007. He joined LIMSI-CNRS in Orsay, where he developed machine translation activities within the " Traitement du Language Parlé " team. He joined the CNRS during his term as Director of LIMSI (2013-2019). His research activities cover a wide spectrum of topics in automatic language processing, from computational morphology to text mining and structured learning methods. In recent years, the focus has been on multilingual processing : machine translation and alignments, interlingual transfer learning, study of large massively multilingual language models.

Speaker(s)

François Yvon

CNRS Research Director