Towards Building a Trustworthy RAG-Based Chatbot for the Italian Public Administration

Building a Trustworthy Retrieval-Augmented Generation (RAG) chatbot for Italy’s public sector presents challenges that go beyond selecting an appropriate Large Language Model. A major issue is the retrieval phase, where Italian text embedders often underperform compared to English and multilingual counterparts, hindering precise identification and contextualization of critical information. Regulatory constraints further complicate matters by disallowing closed source or cloud based models, fo...

Attacking RAG Systems in Multiple Domains with Locally Running and Automatic Procedures

Nowadays, services exploiting Large Language Models (LLMs) frequently customize the model responses in function of their specific requirements, thanks to Retrieval-Augmented Generation (RAG). While this is a simple and effective solution, it has been recently shown to suffer from serious LLM-related security issues that, in some circumstances, can be exploited to access the RAG internal knowledge base, possibly containing private/sensitive data. This paper (i) compares different existing rece...

Sentinel-2 and PRISMA satellite data for urban land classification in Tuscany region (TUS:CAN PROJECT)

The TUS:CAN project, part of the ASI program "Innovation for Downstream Preparation - Public Administrations," focuses on sustainable urban planning through the classification of man-made materials using multispectral and hyperspectral satellite data. The project aims at mapping the land cover of natural and artificial materials in urban and suburban areas using 10m-hypersharpened Sentinel-2 data with a Random Forest classifier. PRISMA hyperspectral data further categorize artificial material...

Dall’Amiata alla Maremma. Ildebrando Imberciadori e la costruzione dell’identità storica del territorio grossetano ai tempi della riforma agraria