Breaking Windows Malware Detection: A Comprehensive Evaluation of Problem-Space Adversarial Robustness

arXiv:2609.34456 · cs.CR · Submitted 2026-09-28 · Read on arXiv

cs.CR

Submitted: 2026-09-28

Updated: 2026-09-28

Code: https://github.com/mashalzainab/windows-malware-adversarialrobustness

Terminology

• Robustness — 65× in this paper · explained in 5 episodes
• Transferability — 40× in this paper · explained in 4 episodes
• Adversarial Examples — 28× in this paper · explained in 2 episodes
• Reinforcement Learning — 13× in this paper · explained in 4 episodes
• Adversarial Attacks — 10× in this paper · explained in 1 episode
• Adversarial Robustness — 10× in this paper · explained in 3 episodes
• Benchmark — 10× in this paper · explained in 2 episodes
• Accuracy — 8× in this paper · explained in 1 episode
• AUC — 7× in this paper · explained in 2 episodes
• Action Space — 4× in this paper · explained in 1 episode
• Generalization — 4× in this paper · explained in 13 episodes
• Attack Success Rate — 3× in this paper
• Deep Reinforcement Learning — 3× in this paper · explained in 2 episodes
• Adversarial Training — 2× in this paper · explained in 1 episode
• Consistency — 2× in this paper · explained in 3 episodes
• Feature Extraction — 2× in this paper · explained in 2 episodes
• IEEE Security and Privacy Workshops — 2× in this paper
• Native Test Set — 2× in this paper
• Provenance — 2× in this paper · explained in 4 episodes
• Unified Evaluation Set — 2× in this paper
• Alignment — 1× in this paper · explained in 1 episode
• Attack Success Rate (ASR) — 1× in this paper · explained in 2 episodes
• Computational overhead — 1× in this paper · explained in 2 episodes
• Core — 1× in this paper · explained in 2 episodes
• Feature Engineering — 1× in this paper · explained in 3 episodes
• Generative Adversarial Networks (GANs) — 1× in this paper · explained in 3 episodes
• IEEE Congress on Evolutionary Computation — 1× in this paper
• Representation Learning — 1× in this paper · explained in 2 episodes
• Transfer Learning — 1× in this paper · explained in 10 episodes
• Transformer — 1× in this paper · explained in 2 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• MAB-Malware: A Reinforcement Learning Framework for Attacking Static Malware Classifiers
• COPYCAT: Practical Adversarial Attacks on Visualization-Based Malware Detection
• On the Effectiveness of Adversarial Training on Malware Classifiers
• Explaining and Harnessing Adversarial Examples
• Explaining Vulnerabilities of Deep Learning to Adversarial Malware Binaries
• Learning to Evade Static PE Machine Learning Malware Models via Reinforcement Learning
• Malware Classification Using Static Disassembly and Machine Learning
• EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models
• Transfer Learning for Image-Based Malware Classification
• Malware Detection by Eating a Whole EXE
• Realizable Universal Adversarial Perturbations for Malware
• EXE-Bench: Ranking the Tradeoffs of AI-based Windows Malware Detectors for Real-World Usability
• On the Security Risks of ML-based Malware Detection Systems: A Survey
• Evaluating the Robustness of a Production Malware Detection System to Transferable Adversarial Attacks
• Towards Deep Learning Models Resistant to Adversarial Attacks
• Interpreting and Evaluating Neural Network Robustness
• SOREL-20M: A Large Scale Benchmark Dataset for Malicious PE Detection
• Deceiving End-to-End Deep Learning Malware Detectors using Adversarial Examples
• MERLIN -- Malware Evasion with Reinforcement LearnINg
• Binary Black-box Evasion Attacks Against Deep Learning-based Static Malware Detectors with Adversarial Byte-Level Language Model

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs