Nick Rahimi

Attack-specific detection strategies, learned through a weighted combination of ensemble disagreement, predictive uncertainty, and distributional anomaly signals, significantly boost adversarial robustness in network intrusion detection.

Fusing BanglaBERT with stacked LSTMs achieves state-of-the-art multi-label cyberbullying detection in Bangla, outperforming unimodal approaches by effectively capturing both contextual nuances and sequential dependencies in text.