people

I am a PhD student in Applied Data Science and Artificial Intelligence at University of Trieste, doing my PhD in collaboration with AREA Science Park. I am currently visiting INRIA in the Datashape team in Sophia-Antipolis, collaborating with Mathieu Carrière.

My research focuses on Topological Data Analysis and Interpretability of Large Language Models. By background I am a Computer Scientist (Master and Bachelor from University of Modena and Reggio Emilia), passionate about multidisciplinary problems.

Recent work:

• Interpretability of LLMs using TDA (ICML’25)
• Topological Data Analysis on neuroscience data
• Complex Systems (Sensors’24)

I am a postdoctoral researcher with a background in Mathematics and Astrophysics. My master’s degree in Mathematics focussed on algebraic topology, whereas during my PhD and Postdoc I did research in cosmology.

I recently joined Area Science Park to apply the skills acquired through the work above to study the interpretability of large ML networks using topological and geometric methods.

Research interests:

• Topological deep learning
• LLM interpretability via topological data analysis

I am completing my Master’s degree in Mathematics for Data Science at University of Trento and am enrolled in a pre-PhD program at Area Science Park. I obtained my Bachelor’s degree in Mathematics from University of Milano-Bicocca and have a background in statistics and stochastic processes.

I am particularly interested in Topological Data Analysis. In my Master’s thesis, I investigated the integration of multiparameter persistence into neural networks to improve their explainability.

Research interests:

• TDA for explainable models by design
• Interpretation of LLMs via topological methods
• Interplay between TDA and stochastic dynamics

I hold a Master’s in Data Management and Curation from Area Science Park and SISSA, which I completed with honors with a thesis titled “Enhancing the Sensorium Dataset for FAIR Neuroscience Research”. I also hold an M.S. in Chemical Sciences and a PhD in Cognitive Neuroscience. For several years, I have investigated cognitive development in human infants and their exceptional learning capacities during the first months of life using non-invasive neuroimaging techniques, such as electroencephalography (EEG) and functional Near Infrared Spectroscopy (fNIRS).

In my Master’s thesis, I applied topological data analysis methods to a publicly available dataset of neural responses recorded from the primary visual cortex of mice.

Recent Research and Academic Interests:

• Neural mechanisms underlying the remarkable learning abilities of human infants
• Adoption of open data management practices
• Development and application of novel analytical techniques for neural data