siming

★ 111 Open GitHub ↗

Behavioral Grammar: detecting adaptive malware via tiny language-model priors and second-order temporal analysis. 0.88M-param self-supervised HIDS — 93% detection vs adaptive adversarial agent at 3.84% onboarding FPR.

Explore Similar Repositories

njrat-config-analyzer : Best RAT Builder Analysis Tool 2026 for Cybersecurity Education & Malware Detection
robust-adv-malware-detection : [IEEE S&P Workshop 2018] "Adversarial Deep Learning for Robust Detection of Binary Encoded Malware" Abdullah Al-Dujaili, Alex Huang, Erik Hemberg, Una-May O’Reilly
cli : Command line interface for the Phylum API
ML-Executable-Malware-Detector : An offline, machine learning-powered static analysis tool that detects malware in Windows executables using LightGBM and the EMBER dataset.
masc : A Web Malware Scanner
// repository documentation