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
Malware-Detection-Using-Machine-Learning : Multi-class malware classification using Deep Learning
adagio : Structural Analysis and Detection of Android Malware
Malware-Detection-using-Deep-Learning : Firstly, we generate images from benign and malware executable files. Secondly, by using deep learning, we train a model to detect malware files. Then, by the trained model, we try to classify a file as malware or not. By using malware images and deep learning, we can detect malware fast since we do not need any static analysis or dynamic analysis.
active-learning : Continuous Learning for Android Malware Detection (USENIX Security 2023)
// repository documentation