Javi G. Gilabert | Barcelona Supercomputing Center

About Me

I am currently a research engineer at the Barcelona Supercomputing Center. My research is centered on investigating added toxicity in Multilingual Neural Machine Translation, utilizing decoder-only architectures for machine translation, enhancing the translation quality of low-resource languages, and improving the interpretability of large language models.

News

• April 2025: Two papers accepted at Machine Translation Summit 2025: Investigating the translation capabilities of Large Language Models trained on parallel data only and Culture-aware machine translation: the case study of low-resource language pair Catalan-Chinese.
• April 2025: I was invited to give a presentation about SalamandraTA model at FAIR - META. Check the slides here: slides
• March 2025: Our paper “MT-LENS: An all-in-one Toolkit for Better Machine Translation Evaluation” was accepted by NAACL 2025 – Demo Track.
• February 2025: We’ve released our multilingual LLM for translation tasks, built on Salamandra-LLM, capable of translating across 37 European languages! Check it out here: salamandraTA-7b-instruct.
• October 2024: Our paper “Training and Fine-Tuning NMT Models for Low-Resource Languages using Apertium-Based Synthetic Corpora” was accepted by WMT2024.
• June 2024: Release of Plume, an open multilingual LLM for Machine Translation.
• June 2024: New paper on Machine Translation for low resource languages.
• June 2024: Presented our work on mitigating added toxicity in Machine Translation (ReSeTOX) at EAMT 2024.
• November 2023: Presented ReSeTOX at Finland AI Day 2023 (event organized by the Finnish Center for Artificial Intelligence FCAI).

Publications

MT-LENS: An all-in-one Toolkit for Better Machine Translation Evaluation

Javier García Gilabert, Carlos Escolano, Audrey Mash, Xixian Liao, Maite Melero

Conference of the North American Chapter of the Association for Computational Linguistics. Demo track. (NAACL), 2025.

PDF Code
Investigating the translation capabilities of Large Language Models trained on parallel data only

Javier García Gilabert, Carlos Escolano, Aleix Sant, Francesca De Luca Fornaciari, Audrey Mash, Xixian Liao, Maite Melero

Machine Translation Summit XX: Research-Technical Track (MT Summit), 2025.

PDF Code
BSC Submission to the AmericasNLP 2024 Shared Task

Javier García Gilabert, Aleix Sant, Carlos Escolano, Francesca De Luca Fornaciari, Audrey Mash, Maite Melero

Workshop on Natural Language Processing for Indigenous Languages of the Americas (AmericasNLP), 2024.

PDF
ReSeTOX: Re-learning attention weights for toxicity mitigation in machine translation

Javier García Gilabert, Carlos Escolano, Marta R. Costa-Jussà

Conference of the European Association for Machine Translation (EAMT), 2024.

PDF Code
Culture-aware machine translation: the case study of low-resource language pair Catalan-Chinese.

Xixian Liao, Carlos Escolano, Audrey Mash, Francesca De Luca Fornaciari, Javier García Gilabert, Miguel Claramunt Argote, Ella Bohman and Maite Melero.

Machine Translation Summit XX: Research-Technical Track (MT Summit), 2025.
Training and Fine-Tuning NMT Models for Low-Resource Languages using Apertium-Based Synthetic Corpora

Aleix Sant, Daniel Bardanca, José Ramom Pichel Campos, Francesca De Luca Fornaciari, Carlos Escolano, Javier Garcia Gilabert, Pablo Gamallo, Audrey Mash, Xixian Liao and Maite Melero.

The Ninth Conference on Machine Translation (WMT), 2024.

PDF