I'm a CNRS postdoctoral researcher in the COATI team, working under the supervision of 3IA chair Emanuele Natale. COATI is a joint team involving Inria, CNRS, and Université Côte d'Azur. My research spans the theory and applications of machine learning, including representation learning, graph neural networks, learning from noisy labels, and model merging. I completed my PhD in Data Science in 2024 at Sapienza University of Rome, working with Prof. Fabrizio Silvestri and the RSTless group.
Research interests
Education
Publications
2026
ICLR 2026
MASS: MoErging through Adaptive Subspace Selection
AISTATS 2026
The Majority Vote Paradigm Shift: When Popular Meets Optimal
AISTATS 2026
Rank Lifting and Random Non-Linear Maps
ECML-PKDD 2026
Subtract the Corruption: Training-Data-Free Corrective Machine Unlearning using Task Arithmetic
2025
CVPR 2025
Task Singular Vectors: Reducing Task Interference in Model Merging
TMLR 2025
Hypergraph Neural Networks through the Lens of Message Passing: A Common Perspective to Homophily and Architecture Design
IJCNN 2025
Link Prediction under Heterophily: A Physics-Inspired Graph Neural Network Approach
UniReps Workshop (NeurIPS)
On Task Vectors and Gradients
2024
Neural Networks 2024
A Topological Description of Loss Surfaces Based on Betti Numbers
IJCNN 2024
Learning with Noisy Labels through Learnable Weighting and Centroid Similarity
Workshop @ ICML 2024
∇τ: Gradient-based and Task-Agnostic Machine Unlearning
2023
Sensors 2023
False Data Injection Impact on High RES Power Systems with Centralized Voltage Regulation Architecture
2022
IJCNN 2022
Newron: A New Generalization of the Artificial Neuron to Enhance the Interpretability of Neural Networks
Talks & Posters
Teaching & Mentoring
Organizing & Awards
Peer Review & Program Committees