# Can Drift-Adaptive Malware Detectors Be Made Robust? Attacks and Defenses Under White-Box and Black-Box Threats

Source: [Can Drift-Adaptive Malware Detectors Be Made Robust? Attacks and Defenses Under White-Box and Black-Box Threats](<https://www.sota2.com/research/paper/69d7327f7d0172140ec76aeb>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-10T05:00:04.898Z

Authors: Adrian Shuai Li, Md Ajwad Akil, Elisa Bertino

Published: 2026-04-08T02:33:02.000Z

## Abstract

Concept drift and adversarial evasion are two major challenges for deploying machine learning-based malware detectors. While both have been studied separately, their combination, the adversarial robustness of drift-adaptive detectors, remains unexplored. We address this problem with AdvDA, a recent malware detector that uses adversarial domain adaptation to align a labeled source domain with a target domain with limited labels. The distribution shift between domains poses a unique challenge: robustness learned on the source may not transfer to the target, and existing defenses assume a fixed distribution. To address this, we propose a universal robustification framework that fine-tunes a pretrained AdvDA model on adversarially transformed inputs, agnostic to the attack type and choice of transformations. We instantiate it with five defense variants spanning two threat models: white-box PGD attacks in the feature space and black-box MalGuise attacks that modify malware binaries via functionality-preserving control-flow mutations. Across nine defense configurations, five monthly adaptation windows on Windows malware, and three false-positive-rate operating points, we find the undefended AdvDA completely vulnerable to PGD (100% attack success) and moderately to MalGuise (13%). Our framework reduces these rates to as low as 3.2% and 5.1%, respectively, but the optimal strategy differs: source adversarial training is essential for PGD defenses yet counterproductive for MalGuise defenses, where target-only training suffices. Furthermore, robustness does not transfer across these two threat models. We provide deployment recommendations that balance robustness, detection accuracy, and computational cost.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Malware Detection on MB-24+ PGD ε=2/255](<https://www.sota2.com/research/sota/malware-detection-on-mb-24-plus-pgd-2-255.md>): Malware Detection; MB-24+ PGD ε=2/255; ASR: 1.2.

- [Malware Detection on MB-24+ PGD ε=8/255](<https://www.sota2.com/research/sota/malware-detection-on-mb-24-plus-pgd-8-255.md>): Malware Detection; MB-24+ PGD ε=8/255; ASR: 12.1.

- [Malware Detection on MB-24+ MalGuise](<https://www.sota2.com/research/sota/malware-detection-on-mb-24-plus-malguise.md>): Malware Detection; MB-24+ MalGuise; ASR: 3.2.

- [Malware Detection on MB-24+ Clean](<https://www.sota2.com/research/sota/malware-detection-on-mb-24-plus-clean.md>): Malware Detection; MB-24+ Clean; Clean TPR: 70.3.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.06599>)

[PDF](<https://arxiv.org/pdf/2604.06599.pdf>)
