Search papers, labs, and topics across Lattice.
3
0
4
5
The overwhelming majority of ICS cybersecurity datasets ignore critical early-stage threats, limiting the effectiveness of intrusion detection research.
Conventional binary metrics can mask critical performance insights, with macro F1 scores plummeting from 85.44% to 37.84% when using behavior-proxy multiclass predictions.
Automating ICS penetration testing with LLMs can translate technical findings into actionable executive reports, bridging the gap between security researchers and industrial operators.
