Italian Language

Research on the Italian language is rapidly advancing, driven by the need for robust natural language processing (NLP) tools in a language with significant cultural and economic importance. Current efforts focus on developing and adapting large language models (LLMs), including both encoder-decoder and decoder-only architectures, for various tasks such as zero-shot classification, named entity recognition, and text generation in specialized domains like biomedicine and legal contexts. This work also includes the creation of new datasets and benchmarks to evaluate model performance and address the challenges of low-resource scenarios, particularly for dialects and older speakers. These advancements are improving the accuracy and efficiency of NLP applications for Italian, impacting fields ranging from public administration to education and healthcare.