Dobici, Elisa (A.A. 2024/2025) AI-powered document retrieval: a multimodal rag-based conversational assistant for the Italian public administration. Tesi di Laurea in Machine learning, Luiss Guido Carli, relatore Giuseppe Francesco Italiano, pp. 100. [Master's Degree Thesis]

PDF (Full text) Restricted to Registered users only Download (6MB) | Request a copy

Abstract/Index

Artificial intelligence & the rise of large language models. Natural language processing (NLP). Large language models (LLMs). Retrieval-augmented-generation (RAG). Overview of RAG architecture. Vector databases. Variants of RAG architecture. Evaluation of RAG systems. Challenges and limitations of RAG systems. State of the art in RAG systems. Materials & environment configuration. Programming environment. Core components. Runtime infrastructure and monitoring. Methodology & framework implementation. Motivation and high-level architecture. Multimodal RAG pipeline and execution flow. Shared stages across document formats. Format-specific adaptations. User interface and orchestration. System evaluation: experiments & results. Test-set evaluation.

References

Bibliografia: pp. 84-86.

Thesis Type:	Master's Degree Thesis
Institution:	Luiss Guido Carli
Degree Program:	Master's Degree Programs > Master's Degree Program in Data Science e Management (LM-91)
Chair:	Machine learning
Thesis Supervisor:	Italiano, Giuseppe Francesco
Thesis Co-Supervisor:	Finocchi, Irene
Academic Year:	2024/2025
Session:	Summer
Deposited by:	Alessandro Perfetti
Date Deposited:	05 Nov 2025 11:31
Last Modified:	05 Nov 2025 11:31
URI:	https://tesi.luiss.it/id/eprint/43630

Downloads

Repository Staff Only

View Item