I study Android and mobile ecosystems, from vulnerable benchmarks and privacy disclosures to LLM-assisted security analysis.
Latest work
TeleGapper
Privacy-policy reliability in Telegram Mini Apps.
About
I am a PhD student in Cybersecurity at the IMT School for Advanced Studies Lucca, working on mobile and Android security.
My research focuses on practical privacy and security problems: vulnerable Android benchmarks, app-store disclosure inconsistencies, Telegram Mini App privacy, and LLM-based security analysis.
Android
Benchmarks and methods for evaluating mobile app vulnerabilities.
Privacy
Measuring gaps between declared and observed data practices.
LLMs
Evaluating semantic reasoning for obfuscation and security tasks.
Selected work
L. Ferrari, M. Ceccato, L. Verderame
arXiv preprint arXiv:2608.13390
TeleGapper is a black-box dynamic analysis framework for Telegram Mini Apps. It captures runtime network traffic, identifies third-party communications, and compares observed data flows against declared privacy information across 278 working Mini Apps.
L. Ferrari, M. Alecci, J. Samhi, T. F. Bissyandé, J. Klein, M. Ceccato, L. Verderame
arXiv preprint arXiv:2606.14233
A large-scale empirical study on whether off-the-shelf LLMs can detect and classify obfuscation in Android apps through semantic reasoning, without handcrafted rules, predefined signatures, or dedicated training.
L. Ferrari, B. Habbati, M. Guerar, M. Ceccato, L. Verderame
arXiv preprint arXiv:2604.16128
PolicyGapper is an LLM-based methodology and automated tool that checks inconsistencies between Google Play Data Safety Sections and privacy policies.
L. Ferrari, F. Pagano, L. Verderame, A. Merlo
IEEE European Symposium on Security and Privacy Workshops (EuroS&PW), pp. 569–580
A comprehensive vulnerable Android app dataset aligned with OWASP MASVS, designed to evaluate Static Application Security Testing tools against modern mobile vulnerability patterns.