Newest papers

Mal-AAEGAN: Adversarial Autoencoder with Double-Critic WGAN-GP for Malware Generation

Nam Hoang Thanh, Trung Pham Duy, Lam Bui Thu · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26488 · cited by 0 works (OpenAlex) · open-access PDF

German

Generative adversarial Netze (GANs) wie MalGAN, Mal-LSGAN und unsere vorherigen gemeinsamen Mal-D2GAN können adversarial malware erstellen, um diese Detektoren zu erforschen und zu härten, aber sie arbeiten direkt auf einem hochdimensionalen, diskreten binären Feature-Raum, was zu instabilem Training, verschwinden Gradienten und adversarial Basismuster von begrenzter Vielfalt führt. Um diese Probleme anzugehen, schlagen wir Mal-AAEGAN vor. Erstens, ein adversarial autoencoder (AAE) erlernt eine niedrige Dimension, reguliert und kontinuierlich latente Darstellung des $M$-dimensionalen binären Feature-Raums, und der Generator führt zu einer additiven Störung in diesem latenten Raum und nicht im Rohraum. Zweitens folgt das Training

Machine translation (Meta M2M100); the English original follows.

Abstract

Machine-learning (ML) malware detectors are known to be vulnerable to adversarial samples. Generative adversarial networks (GANs) such as MalGAN, Mal-LSGAN, and our previous Mal-D2GAN can craft adversarial malware to probe and harden these detectors, yet they operate directly on a high-dimensional, discrete binary feature space, which leads to unstable training, vanishing gradients, and adversarial samples of limited diversity. To address these problems we propose Mal-AAEGAN. First, an adversarial autoencoder (AAE) learns a low-dimensional, regularized and continuous latent representation of the $M$-dimensional binary API feature vector, and the generator performs an additive perturbation in this latent space rather than in the raw binary space. Second, training follows the Wasserstein GAN with gradient penalty (WGAN-GP) scheme using two critics: a mimic-black-box critic that inherits the role of the substitute detector and a naive critic that inherits the role of the additional detector. A decode-binarize-OR pipeline guarantees that the original API calls of the malware are preserved. The model is evaluated on two malware datasets of different input dimensionality ($M\in\{139,215\}$) against eight common ML detectors and compared, under an identical protocol, with MalGAN, Mal-LSGAN and Mal-D2GAN. Mal-AAEGAN trains stably in the continuous latent space and attains an evasion rate on par with or higher than these binary-space baselines across both datasets, while its generator stays responsive to the noise input rather than reducing to a single fixed perturbation per malware.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Malware, Adversarial system, Autoencoder, Artificial intelligence, Discriminator

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange