---
title: README
emoji: 💻
colorFrom: gray
colorTo: yellow
sdk: static
pinned: false
---
This is the official space for the paper: "Optimizing LLMs for Italian: Reducing Token Fertility and Enhancing Efficiency Through Vocabulary Adaptation"
[](https://2025.naacl.org/)
[](https://arxiv.org/abs/2504.17025v1)
[](https://creativecommons.org/licenses/by-nc-sa/4.0/)
[](https://huggingface.co/collections/SemanticAlignment/mistral-7b-v01-adapted-679243206cec8a21f75435dd)
[](https://huggingface.co/collections/SemanticAlignment/llama-31-adapted-67924314d8957c78a3e7bcaf)
This space is shared between three Italian institutions that led the work:
- [SapienzaNLP](https://nlp.uniroma1.it/) -- Sapienza university of Rome
- [Istituto di Scienza e Tecnologie dell'Informazione "A. Faedo"](https://www.isti.cnr.it/it/) -- CNR Pisa
- [Istituto di Linguistica Computazionale "A. Zampolli"](https://www.ilc.cnr.it/) -- CNR Pisa