Lotfi

The paper proposes a novel semi-automated method to perform continuous threat modeling by inferring the actual system architecture from combined static configuration and dynamic network flow data, significantly improving threat detection over isolated component analysis.

The paper proposes an attention-guided hybrid framework combining 1D and 2D CNNs to robustly enhance ECG-based biometric recognition, achieving high accuracy across multiple datasets and demonstrating stable performance over long periods.

The paper demonstrates that clinical vision-language models (VLMs) pose a significant privacy risk by allowing de-identified images to be re-linked to original reports, and proposes a targeted differential privacy (DP) fine-tuning method to mitigate this risk while preserving clinical utility.