Davide Maiorca

The paper proposes a model-agnostic framework to evaluate combining Active Learning (AL) and Semi-Supervised Learning (SSL) techniques for malware detection, demonstrating that these combined methods can reduce manual labeling costs by up to 90% while maintaining high detection performance.

This paper empirically demonstrates that current Static Application Security Testing (SAST) tools are fundamentally unreliable against common JavaScript obfuscation techniques, showing that obfuscation can lead to near-total evasion of vulnerability detection.