Multilingual Generative Language Models with a focus on the Italian language: training and evaluation

The selected candidate will conduct research in the field of Natural Language Processing (NLP), with a particular focus on the training, evaluation, and improvement of multilingual language models. The candidate will contribute to the development of the Minerva project, a language model trained from scratch with a primary focus on the Italian language, and will investigate novel methodologies for the collection, curation, and cleaning of training data for generative language models, as well as for their evaluation. The research will focus on developing techniques to improve both learning and evaluation in knowledge domains that are underrepresented in current state-of-the-art language models.