# Empirical Quantification of Spurious Correlations in Malware Detection

arXiv: https://arxiv.org/abs/2506.09662 · HTML: https://mlsanity.com/paper/2506.09662/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 14 references that are in the map.
- First posted on arXiv 11 Jun 2025.

## Builds on (most cited first, 10 of 10)

- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Language Modeling with Gated Convolutional Networks](https://mlsanity.com/paper/1612.08083/) (2016)
- [EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models](https://mlsanity.com/paper/1804.04637/) (2018)
- [Malware Detection by Eating a Whole EXE](https://mlsanity.com/paper/1710.09435/) (2017)
- [Dos and Don'ts of Machine Learning in Computer Security](https://mlsanity.com/paper/2010.09470/) (2020)
- [Functionality-preserving Black-box Optimization of Adversarial Windows Malware](https://mlsanity.com/paper/2003.13526/) (2020)
- [Explaining Vulnerabilities of Deep Learning to Adversarial Malware Binaries](https://mlsanity.com/paper/1901.03583/) (2019)
- [Activation Analysis of a Byte-Based Deep Neural Network for Malware Classification](https://mlsanity.com/paper/1903.04717/) (2019)
- [Quo Vadis: Hybrid Machine Learning Meta-Model based on Contextual and Behavioral Malware Representations](https://mlsanity.com/paper/2208.12248/) (2022)
- [SLIFER: Investigating Performance and Robustness of Malware Detection Pipelines](https://mlsanity.com/paper/2405.14478/) (2024)

## Built on by (most cited first, 1 of 1)

- [Beyond the Wrapper: Identifying Artifact Reliance in Static Malware Classifiers using TRUSTEE](https://mlsanity.com/paper/2605.07034/) (2026)

## Abstract (arXiv)

End-to-end deep learning exhibits unmatched performance for detecting malware, but such an achievement is reached by exploiting spurious correlations -- features with high relevance at inference time, but known to be useless through domain knowledge. While previous work highlighted that deep networks mainly focus on metadata, none investigated the phenomenon further, without quantifying their impact on the decision. In this work, we deepen our understanding of how spurious correlation affects deep learning for malware detection by highlighting how much models rely on empty spaces left by the compiler, which diminishes the relevance of the compiled code. Through our seminal analysis on a small-scale balanced dataset, we introduce a ranking of two end-to-end models to better understand which is more suitable to be put in production.