Researchers have developed ViPER, a novel approach to malware detection that utilizes image-based analysis combined with an awareness of executable packing. This method employs a Vision Transformer (ViT) backbone adapted with LoRA and a dual-head architecture to simultaneously classify malware and detect packing. A packing-aware gating mechanism allows for distinct predictions based on whether a binary is packed, addressing a key challenge where packed files appear as high-entropy images. ViPER achieved strong performance on a dataset of 200,000 Windows PE byteplot images, with a balanced accuracy of 0.8521 and an ROC-AUC of 0.9260 for malware detection, alongside a packing detection AUC of 0.9949. AI
RANK_REASON The cluster contains an academic paper detailing a new method for malware detection. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →