Oral presentation at CLiC-it 2024
We evaluate the rebus-solving capabilities of large language models on a new Italian dataset.
IT5s are the first encoder-decoder transformers pretrained on more than 40 billion Italian words.
We present Inseq, a Python library to democratize access to interpretability analyses of sequence generation models.
An open-source library to democratize access to model interpretability for sequence generation models
We present the first CLIP model for the Italian Language (CLIP-Italian), trained on more than 1.4 million image-text pairs.
The first CLIP model pretrained on the Italian language.