Mathieu Blondel This is actually not new. We (with Karteek Alahari) published this result at NeurIPS 2021 in our paper on Frank-Wolfe for MAP inference (proceedings.neurips.cc/paper/2021/has…), in which we provide convergence results for the general non-convex and non-smooth settings. 1/2

In our NeurIPS 2021 paper (with Karteek Alahari) we showed that CCCP is Frank-Wolfe in disguise. Happy to see other people recently rediscovering this fact and presenting it as a striking result. Want to know another equally striking fact? Mean Field is also Frank-Wolfe!👇

_Jamais 2 sans 3_: we are thrilled to announce the 3⃣rd edition of the NLP winter School 🏔️ALPS🏔️: hear and interact with leading researchers and ~ 100 other NLP students Date: 16-20th Jan 2023 Application Deadline: Sept 16th alps.imag.fr

We just opened a PhD Position (between NAVER LABS Europe and MT Group at FBK ) to start in Nov 2022. PhD topic: Unified Foundation models for Speech-to-Speech Translation. Application deadline: August 23rd. More details here: tinyurl.com/PhD-FBK-NLE . Please RT (or DM to know more)

Missed last INTERSPEECH 2025 conference ? Attended remotely and got lost in gathertown ? I wrote my highlights of the conference in this blog: europe.naverlabs.com/blog/highlight… Very biased to my current personal interests indeed ! #interspeech2022

Pre-training for Speech Translation: CTC Meets Optimal Transport abs: arxiv.org/abs/2301.11716

Many papers recently exploit optimal transport-based losses, and it is great to see how this can be done to improve ST models in arxiv.org/abs/2301.11716. I would have been curious to see results on the Conformer model and with CTC compression!

Our Pick of the week: Phuong-Hang Le et al., "Pre-training for Speech Translation: CTC Meets Optimal Transport" by Marco Gaido arxiv.org/abs/2301.11716 #NLProc #optimaltransport #CTC #speechtranslation

Durant la Journée Journée R&T Cognition 2023 journee-rt-cognition.fr, didier_schwab vous accueillera en présentant GAZEPLAY-EVAL : un outil générique pour mettre en place des test d’évaluation des capacités langagières. Stand S2-11

Prof. Gil Strang's last lecture is ... Monday! Watch it via live stream: grinfeld.org/strang/

A nice combination of strong techniques for speech translation, including an interesting extension of our Siamese pre-training!

We are doing🏔️ALPS 🏔️again, 4⃣th edition A week full of insightful lectures, knowing fellow researchers and dreaming of joint projects ≠ this time: we actually meet in the Alps. Close to a French National Park, at 1500m, surrounded by mountains alps.imag.fr

Presented this week at ACL (from NLP@NLE): Memory-efficient NLLB-200: language-specific expert pruning of a massively multilingual machine translation model - Naver Labs Europe europe.naverlabs.com/research/publi…

Presented this week at IWSLT@ACL: NAVER LABS Europe’s Multilingual Speech Translation Systems for the IWSLT2023 Low-Resource Track (were ranked 1st on the 2 lang pairs they participated in this track) - aclanthology.org/2023.iwslt-1.1… with @mzboito I.calapodescu, a.berard & e.gow-smith !

Projet ANR Pantagruel accepté :) Construction et évaluation de giga modèles de langue multimodaux et inclusifs (écrit, oral, pictogrammes) pour le français général et clinique anr.fr/fileadmin/aap/…

Congratulations to the team on this exciting release featuring an all-in-one massively multilingual multimodal machine translation model! What I'm particularly excited about for this work👇

Opt for lang-specific encoders for real-world applications in lang like French. LeBenchmark 2.0 is here with 3 new pre-trained speech SSL models (until 1B), tested on 6 tasks, powered by Jean Zay #HPC from Genci. Paper: arxiv.org/abs/2309.05472 Models: huggingface.co/LeBenchmark

SeamlessStreaming is an AI translation model that can deliver state-of-the-art results on streaming translation with <2 seconds of latency. One core piece of our latest Seamless Communication research work by teams at FAIR. More on this project ➡️ bit.ly/4165c9z