Dissertation topics suggested by the teacher.
Proposals of master theses in Artificial Intelligence are occasionally published on the
Language Technologies Lab website > For Students
Recent dissertations supervised by the teacher.
Second cycle degree programmes dissertations
- A Comparative Study of Multimodal Deep Learning Architectures for Depression Detection
- A Comparison between LLMs and SLMs for Document Processing in the Insurance Sector
- A Multimodal Approach to Alternative Text Generation for Accessibility using Large Language Models and Computer Vision Classifiers
- A Self-Supervised Attribution Method for Explaining Neural Networks
- Agentic and Privacy-Preserving Retrieval-Augmented Generation: Architecture and Empirical Evaluation
- AI solutions for Visual Recognition of Thyroid Eye Disease
- An agentic RAG approach for knowledge management in enterprise consulting contexts
- An Artificial Intelligence System for the Autonomous Pagination of Newspapers
- An LLM Agent System for Reducing Hallucinations in Drug Discovery Tasks
- An LLM-as-a-judge Comparison of Embedding Models for RAG Tasks in Italian Press News Data
- Assessing LLM Capabilities in Low-Resource Language Settings: A Kazakh Perspective
- Balancing Fairness and Semantic Integrity in Sentence Embeddings: A Neuro-Symbolic Approach to Gender Debiasing
- Bridging the Lexical-Semantic Gap in Music Discovery: The Reranker as a Foundational Enabler for Hybrid Agentic RAG Architectures
- Design and Development of a RAG-Based Chatbot System for Web Content Exploration
- Design and Implementation of a High-Accuracy Text-to-SQL Pipeline with LLMs
- Design, Implementation and Benchmarking of a Retrieval-Augmented Chatbot for the Insurance Sector
- Development of an LLM-based system for the generation of multiple-choice English grammar exercises
- Dynamic Few-Shot Learning for Knowledge Graph Question Answering
- Enhancing Document Parsing and Question Answering through Optimized Table Parsing
- Enhancing the Software Development Life Cycle with Conversational AI: LLM-Driven Documentation Q&A and Automated Model Evaluation
- Exploring Knowledge Distillation in LLMs
- Extracting Organization-Location Relations from German News Articles: A Comparative Study of Techniques
- From Knowledge Extraction to Semantic Reasoning: A Comparative Analysis of LLMs
- Graph Orchestrated Multi-Agent Workflows for Insurance Client Intake
- Implementing large language model-based machine translation in small and medium-sized enterprises
- Integrating behaviour trees and finite state machines for human-robot interaction
- Investigating Mechanistic Interpretability in Large Language Models through Model Stitching Across Architectures
- Judging the Judges: A Case Study on LLM-as-a-Judge for Retrieval-Augmented Generation
- Leveraging Fine-Tuned LLMs for Policy Reasoning in Identity and Access Management
- Leveraging Large Language Models for Fact-Checked Claim Retrieval Based On Similarity
- Leveraging LLMs as noisy label generators for named entity recognition
- Machine Learning Approaches for State of Health Estimation of Electric Vehicle Batteries
- Named Entity Recognition for Historical Italian Texts: Overcoming Data Limitations through Strategic Annotation and Model Adaptation
- Optimizing Listwise Reranking with ModernBERT: A Data-Centric Approach with a Novel Graded Corpus
- Progettazione e implementazione di un chatbot per l'assistenza clienti di un servizio di telepedaggio
- Prompt Sensitivity to Context in LLM Multi-Agent Decision-Making: a Prisoner’s Dilemma Case Study
- Selective rationalization via genetic-based learning: extending the GenSPP architecture with the CMA-ES algorithm
- Self-Assessment in RAG Systems: Evaluating Feedback Loops and Quality Control Mechanisms
- Thinking with Images in Large Multimodal Reasoning Models
- Ulisses x AI: Underwater Targets Detection using DIFAR Sonobuoys
- Valida: AI and NLP Techniques in Document Forgery Detection
- VerifAi: Towards an Open-Source Scientific Generative Question-Answering System with Referenced and Verifiable Answers
- Zero-Shot Warning Generation for Misinformative Multimodal Content Detection
