- LiLowLa (2022-2025): Lightweight neural translation technologies for low-resource languages. The main objectives of the project are: The development of a smart crawling method able to prioritize the most productive websites; the development of data augmentation techniques for training neural machine translation systems for low-resource languages; to devise a method for distilling the translation knowledge encoded in large pre-trained models; to enable translation memory-based computer-aided translation tools to exploit target-language monolingual corpora; and to deepen the understanding of how NMT systems behave at prediction time and during training. Project website.
- MaCoCu (2021-2023): Massive collection and curation of monolingual and bilingual data for the under-resourced languages of the Europe Union. Project website.
- MultitraiNMT (2019-2022): Machine translation training for multilingual citizens. The project aims at developing an innovative syllabus in machine translation, and in particular machine translation based on currently popular deep learning techniques, i.e., neural machine translation. Project website.
- GoURMET (2019-2022): The aim of GoURMET is to use and improve neural machine translation for low-resource language pairs and domains. It has five objectives: (i) Advancing low-resource deep learning for natural language applications; (ii) develop a high-quality machine translation for low-resource language pairs and domains; (iii) develop tools for media analysts and journalists; (iv) create a sustainable and maintainable platform and services; and (v) inform stakeholders and user group of project results. It was funded by the EU; grant agreement number 825299. Project website.
Transducens
Research on machine translation, digital libraries and computer-assisted education