Ph.D. candidate in Ingegneria Informatica E Dei Sistemi , 40th cycle (2024-2027)
Department of Control and Computer Engineering (DAUIN)
External Collaborator
External Collaborator
Students and Education (STUDI)
Docente esterno e/o collaboratore didattico
Department of Control and Computer Engineering (DAUIN)
Ospite / Studente in missione
Department of Control and Computer Engineering (DAUIN)
Profile
PhD
Research topic
Machine Unlearning
Tutors
Research interests
Biography
- Bachelor’s degree in Computer Engineering @ PoliTO
- Master’s Degree in Data Science and Engineering @ PoliTO
- Team Leader of the team MALTO (MAchine Learning @ PoliTO)
- Master Thesis: “Bad Teaching in Machine Unlearning with Similarity-based Sampling”
- PhD student @DAUIN in the DBDMG group
- Supervisor: Prof.ssa Elena Maria BARALIS
- Co-Supervisor: Dr. Flavio GIOBERGIA
- PhD Main Topic: Machine Unlearning:
- Remove the influence of selected information from a model’s training.
- Applicable to various model types (e.g., neural networks, transformers, decision trees) and data modalities (e.g., text, images, audio).
- There are many reasons for Machine Unlearning, like complying with new AI regulations (such as GDPR or AI Act), recovery of attacked models, removing biases, copyright claims…
- Other topics of interest: Large Language Models, Explainable AI
Teaching
Teachings
Master of Science
- Large Language Models for Software Engineering. A.A. 2025/26, INGEGNERIA INFORMATICA (COMPUTER ENGINEERING). Collaboratore del corso
- Data science and machine learning lab. A.A. 2025/26, DATA SCIENCE AND ENGINEERING. Collaboratore del corso
- Business intelligence per big data. A.A. 2024/25, INGEGNERIA GESTIONALE. Collaboratore del corso
- Distributed architectures for big data processing and analytics. A.A. 2024/25, DATA SCIENCE AND ENGINEERING. Collaboratore del corso
Research
Research groups
Publications
Latest publications View all publications in Porto@Iris
- Giobergia, Flavio; Savelli, Claudio; Koudounas, Alkis; Baralis, Elena Maria (2025)
Quantum feature selection from interpretable models using qubo formulation. In: CLEF 2025, Madrid (ESP), 9 – 12 September 2025
Contributo in Atti di Convegno (Proceeding) - Savelli, Claudio; Koudounas, Alkis; Giobergia, Flavio (2025)
MALTO at SemEval-2025 task 3: Detecting hallucinations in LLMs via uncertainty quantification and larger model validation. In: 19th International Workshop on Semantic Evaluation (SemEval-2025), Vienna (AT), July 31 - August 1, 2025, pp. 1318-1324. ISBN: 979-8-89176-273-2
Contributo in Atti di Convegno (Proceeding) - Savelli, Claudio; Munis, Evren; Bayat, Erfan; Grieco, Andrea; Giobergia, Flavio (2025)
MALTO at SemEval-2025 Task 4: Dual Teachers for Unlearning Sensitive Content in LLMs. In: 19th International Workshop on Semantic Evaluation (SemEval-2025), Vienna (AT), July 31 - August 1, 2025, pp. 1747-1752. ISBN: 979-8-89176-273-2
Contributo in Atti di Convegno (Proceeding) - Koudounas, Alkis; Savelli, Claudio; Giobergia, Flavio; Baralis, Elena (2025)
"Alexa, can you forget me?" Machine Unlearning Benchmark in Spoken Language Understanding. In: Interspeech 2025, Rotterdam (NL), 17-21 August, 2025
Contributo in Atti di Convegno (Proceeding) - Bertetto, Lorenzo; Bettinelli, Francesca; Buda, Alessio; Da Mommio, Marco; DI BARI, ... (2024)
Towards an Explorable Conceptual Map of Large Language Models. In: 36th International Conference on Advanced Information Systems Engineering, CAiSE 2024, Limassol (CYP), June 3–7, 2024, pp. 82-90. ISSN 1865-1348. ISBN: 9783031609992
Contributo in Atti di Convegno (Proceeding)