A strictly additive CVAE framework injects 20 API imports to reduce malware recall from 87.5% to 30% while routing 99% of evaded samples into a pre-selected benign category, with transfer shown on VirusTotal.
Adversarial malware sample generation method based on the prototype of deep learning detector,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning to Look Benign: Targeted Evasion of Malware Detectors via API Import Injection
A strictly additive CVAE framework injects 20 API imports to reduce malware recall from 87.5% to 30% while routing 99% of evaded samples into a pre-selected benign category, with transfer shown on VirusTotal.