Marco Di Gennaro

Cybersecurity PhD Student at @polimi, 26, Italy.

[Gabriele Digregorio @Io_no__]

📣 Our paper “On the (In)Security of Loading Machine Learning Models” has been accepted at IEEE S&P 2026 (13% acceptance rate this cycle) (1/5)

👇Preprint: https://t.co/YTJsBUw460

#ieeesp #ieeesp26 #ieee #softwaresecurity #aisecurity #cybersecurity #machinelearning #ml #zeroday

[Gabriele Digregorio @Io_no__]

We're researching how people load and share ML models. If that's you, we'd really appreciate your input!

Fill out our quick (1-2 minutes), anonymous survey: https://t.co/CQR7UjzaJK

#ML #AI #MachineLearning

[Marco Di Gennaro @marcodige8]

This research was developed at the @necstlaboratory (@polimi), in the context of the @TRUSTroke_EU project, in collaboration with my co-authors: Giovanni De Lucia, @aScarecrowHat , @raistolo, and @mcarminati1988.

[Marco Di Gennaro @marcodige8]

🚨 Our PoPETS 2025 work, "TimberStrike," finds that federated tree models are vulnerable to privacy leakage via dataset reconstruction.

See you in Washington D.C. in July!

📄 Preprint: https://t.co/Ssaj4521lT
#FederatedLearning #Privacy #PoPETS

[Marco Di Gennaro @marcodige8]

We found that federated tree-based models (like XGBoost) are vulnerable to dataset reconstruction attacks, and we demonstrated this vulnerability across major FL frameworks. Our work introduces a novel attack and provides guidelines for building stronger defenses.