Research projects

TraLaLaM (Translating with Large Language Models)

ressources.chapsvision.com/fr/tralalam/projet ↗

The aim of the project is to explore the use of large language models (LLMs) for machine translation, by asking two main questions: (i) in what scenarios can contextual information be effectively used via prompting? and (ii) for low-resource scenarios (with a focus on dialects and regional languages), can LLMs be effectively trained without any parallel data?

The project is led by Josep Crego (Systran). My role in the project is leader for Inria and head of the work package on translation for low-resource translation without parallel data.

MaTOS (Machine Translation for Open Science)

anr-matos.github.io ↗

The aim of the project is to develop new methods for the machine translation (MT) of complete scientific documents, as well as automatic metrics to evaluate the quality of these translations. Our main application target is the translation of scientific articles between French and English, where linguistic resources can be exploited to obtain more reliable translations, both for publication purposes and for gisting and text mining.

The project is led by François Yvon (CNRS). My role in the project is leader for Inria and head of the work package on the evaluation of the MT for scientific documents.

CoLAF (Corpus et Outils pour les Langues de France 'Corpora and Tools for the Languages of France'

colaf.huma-num.fr ↗

The Inria DEFI COLaF (Corpus and Tools for the Languages of France) aims to provide open-source datasets and tools for automatic text and speech processing for the languages and speakers of France.

The project is led by Benoît Sagot (ALMAnaCH team) and Slim Ouni (MULTISPEECH team).

PR[AI]RIE-PSAI

prairie-psai.fr ↗

PR[AI]RIE-PSAI (Paris School of AI), a follow-up to the PRAIRIE institute, is the largest of the AI Clusters established as part of the France 2030 national strategy. Led by PSL University, the project aims to create an internationally renowned school specialising in artificial intelligence. Its goal is to advance knowledge in AI, provide world-class higher education, and produce groundbreaking innovations in this field. My role in the project is fellow.

DAdaNMT

The aim of this project is to investigate domain adaptation for neural machine translation. We will be exploring the adaptation of models to specific, low-resource domains domains as well as training models for multiple domains.

My role in the project was PI.

BigScience

bigscience.huggingface.co ↗

The BigScience collaboration was a huge community project centred around the training of the open-source large language model, BLOOM.)

My role in the project was as a main contributor to the evaluation work package, especially for the machine translation task.

PRAIRIE

anr.fr/ProjetIA-19-P3IA-0001 ↗

The PRAIRIE Institute (PaRis AI Research InstitutE) was one of the four French Institutes of Artificial Intelligence, which were created as part of the national French initiative on AI announced by President Emmanuel Macron on May 29, 2018. PRAIRIE brought together academic members (“PRAIRIE chairs”) who excel at research and education in both the core methodological areas and the interdisciplinary aspects of AI, and industrial members that are major actors in AI at the global level and a very strong group of international partners.

My role in the project was holder of a Junior chair position.