Simone Aonzo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Simone Aonzo has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Davide Balzarotti
EURECOM
7 shared publications - Savino Dambra
GenDigital, Biot, France
5 shared publications - Alessio Merlo
CASD - School of Advanced Defense Studies, Rome, Italy
3 shared publications - Antonio Ruggia
DIBRIS -University of Genoa Via Dodecaneso 35 I-16146 Genoa Italy
3 shared publications - Andrea Oliveri
EURECOM, Biot, France
2 shared publications - Andrea Possemato
EURECOM, Sophia Antipolis, France
2 shared publications
Latest Publications
- 23 October 2026
- Published by Springer Nature
- p. 475-495
Beyond Isolated Phishing Emails: Discovering Hidden Campaign Relationships with MCDA
- 23 October 2026
- Published by Springer Nature
- p. 475-495
Beyond Isolated Phishing Emails: Discovering Hidden Campaign Relationships with MCDA
- Open Access
- 21 October 2026
- Published by Elsevier in Computers & Security
- Vol. 173
A hybrid static-dynamic analysis for identifying BYOVD vulnerabilities
- 21 October 2026
- Published by Elsevier in Computers & Security
- Vol. 173
Open AccessA hybrid static-dynamic analysis for identifying BYOVD vulnerabilities
- Open Access
- 4 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1371-1386
SoK: Systematization, Detection, and Hunting of Windows Malware Persistence Techniques
- 4 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1371-1386
Open AccessSoK: Systematization, Detection, and Hunting of Windows Malware Persistence Techniques
- 5 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 3575-3587
Trust Under Siege: Label Spoofing Attacks Against Machine Learning for Android Malware Detection
Machine Learning (ML) malware detectors rely heavily on crowd-sourced AntiVirus (AV) labels, with platforms like VirusTotal serving as trusted sources of malware annotations. But what if attackers could manipulate these labels to classify benign software as malicious? We introduce label spoofing attacks, a new threat that contaminates crowd-sourced datasets by embedding minimal and undetectable malicious patterns into benign samples. These patterns coerce AV engines into misclassifying legitimate files as harmful, enabling poisoning attacks against ML-based malware classifiers trained on those data. We demonstrate this scenario by developing AndroVenom, a methodology for polluting realistic data sources and launching subsequent poisoning attacks against ML malware detectors. Experiments show that not only are state-of-the-art feature extractors unable to filter such injections, but various ML models experience Denial-of-Service (DoS) with as little as 1% poisoned samples. Additionally, attackers can flip decisions for specific unaltered benign samples by modifying only 0.015% of the training data, threatening their reputation and market share, while evading anomaly detectors operating on the training data. We conclude by raising concerns about the trustworthiness of ML training processes based on AV annotations and argue that further investigation is needed to develop more reliable labeling strategies.
- 5 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 3575-3587
Trust Under Siege: Label Spoofing Attacks Against Machine Learning for Android Malware Detection
Machine Learning (ML) malware detectors rely heavily on crowd-sourced AntiVirus (AV) labels, with platforms like VirusTotal serving as trusted sources of malware annotations. But what if attackers could manipulate these labels to classify benign software as malicious? We introduce label spoofing attacks, a new threat that contaminates crowd-sourced datasets by embedding minimal and undetectable malicious patterns into benign samples. These patterns coerce AV engines into misclassifying legitimate files as harmful, enabling poisoning attacks against ML-based malware classifiers trained on those data. We demonstrate this scenario by developing AndroVenom, a methodology for polluting realistic data sources and launching subsequent poisoning attacks against ML malware detectors. Experiments show that not only are state-of-the-art feature extractors unable to filter such injections, but various ML models experience Denial-of-Service (DoS) with as little as 1% poisoned samples. Additionally, attackers can flip decisions for specific unaltered benign samples by modifying only 0.015% of the training data, threatening their reputation and market share, while evading anomaly detectors operating on the training data. We conclude by raising concerns about the trustworthiness of ML training processes based on AV annotations and argue that further investigation is needed to develop more reliable labeling strategies.
- 18 October 2025
- Published by Springer Nature
- p. 505-525
The Polymorphism Maze: Understanding Diversities and Similarities in Malware Families
- 18 October 2025
- Published by Springer Nature
- p. 505-525
The Polymorphism Maze: Understanding Diversities and Similarities in Malware Families
- Open Access
- 22 February 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Privacy and Security
- Vol. 28 (2) , 1-33
The Dark Side of Native Code on Android
From a little research experiment to an essential component of military arsenals, malicious software has constantly been growing and evolving for more than three decades. On the other hand, from a negligible market share, the Android operating system is nowadays the most widely used mobile operating system, becoming a desirable target for large-scale malware distribution. While scientific literature has followed this trend, one aspect has been understudied: the role of native code in malicious Android apps. Android apps are written in high-level languages, but thanks to the Java Native Interface (JNI), Android also supports calling native (C/C++) library functions. While allowing native code in Android apps has a strong positive impact from a performance perspective, it dramatically complicates its analysis because bytecode and native code need different abstractions and analysis algorithms, and they thus pose different challenges and limitations. Consequently, these difficulties are often (ab)used to hide malicious payloads. In this work, we propose a novel methodology to reverse engineering Android apps focusing on suspicious patterns related to native components, i.e., surreptitious code that requires further inspection. We implemented a static analysis tool based on such methodology, which can bridge the “Java” and the native worlds and perform an in-depth analysis of tag code blocks responsible for suspicious behavior. These tags benefit the human facing the reverse engineering task: they clearly indicate which part of the code to focus on to find malicious code. Then, we performed a longitudinal analysis of Android malware over the past 10 years and compared the recent malicious samples with actual top apps on the Google Play Store. Our work depicts typical behaviors of modern malware, its evolution, and how it abuses the native layer to complicate the analysis, especially with dynamic code loading and novel anti-analysis techniques. Finally, we show a use case for our suspicious tags: we trained and tested a machine learning algorithm for a binary classification task. Even if suspicious does not imply malicious, our classifier obtained a remarkable F1-score of 0.97, showing that our methodology can be helpful to both humans and machines.
- 22 February 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Privacy and Security
- Vol. 28 (2) , 1-33
Open AccessThe Dark Side of Native Code on Android
From a little research experiment to an essential component of military arsenals, malicious software has constantly been growing and evolving for more than three decades. On the other hand, from a negligible market share, the Android operating system is nowadays the most widely used mobile operating system, becoming a desirable target for large-scale malware distribution. While scientific literature has followed this trend, one aspect has been understudied: the role of native code in malicious Android apps. Android apps are written in high-level languages, but thanks to the Java Native Interface (JNI), Android also supports calling native (C/C++) library functions. While allowing native code in Android apps has a strong positive impact from a performance perspective, it dramatically complicates its analysis because bytecode and native code need different abstractions and analysis algorithms, and they thus pose different challenges and limitations. Consequently, these difficulties are often (ab)used to hide malicious payloads. In this work, we propose a novel methodology to reverse engineering Android apps focusing on suspicious patterns related to native components, i.e., surreptitious code that requires further inspection. We implemented a static analysis tool based on such methodology, which can bridge the “Java” and the native worlds and perform an in-depth analysis of tag code blocks responsible for suspicious behavior. These tags benefit the human facing the reverse engineering task: they clearly indicate which part of the code to focus on to find malicious code. Then, we performed a longitudinal analysis of Android malware over the past 10 years and compared the recent malicious samples with actual top apps on the Google Play Store. Our work depicts typical behaviors of modern malware, its evolution, and how it abuses the native layer to complicate the analysis, especially with dynamic code loading and novel anti-analysis techniques. Finally, we show a use case for our suspicious tags: we trained and tested a machine learning algorithm for a binary classification task. Even if suspicious does not imply malicious, our classifier obtained a remarkable F1-score of 0.97, showing that our methodology can be helpful to both humans and machines.
- Open Access
- 30 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 130-146
How to Train your Antivirus: RL-based Hardening through the Problem Space
ML-based malware detection on dynamic analysis reports is vulnerable to both evasion and spurious correlations. In this work, we investigate a specific ML architecture employed in the pipeline of a widely-known commercial antivirus, with the goal to harden it against adversarial malware. Adversarial training, the most reliable defensive technique that can confer empirical robustness, is not applicable out of the box in this domain, for the principal reason that gradient-based perturbations rarely map back to feasible problem-space programs. We introduce a novel Reinforcement Learning approach for constructing adversarial examples, a constituent part of adversarially training a model against evasion. Our approach comes with multiple advantages. It performs modifications that are feasible in the problem-space, and only those; thus it circumvents the inverse mapping problem. It also makes it possible to provide theoretical guarantees on the robustness of the model against a well-defined set of adversarial capabilities. Our empirical exploration validates our theoretical insights, where we can consistently reach 0% Attack Success Rate after a few adversarial retraining iterations.
- 30 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 130-146
Open AccessHow to Train your Antivirus: RL-based Hardening through the Problem Space
ML-based malware detection on dynamic analysis reports is vulnerable to both evasion and spurious correlations. In this work, we investigate a specific ML architecture employed in the pipeline of a widely-known commercial antivirus, with the goal to harden it against adversarial malware. Adversarial training, the most reliable defensive technique that can confer empirical robustness, is not applicable out of the box in this domain, for the principal reason that gradient-based perturbations rarely map back to feasible problem-space programs. We introduce a novel Reinforcement Learning approach for constructing adversarial examples, a constituent part of adversarially training a model against evasion. Our approach comes with multiple advantages. It performs modifications that are feasible in the problem-space, and only those; thus it circumvents the inverse mapping problem. It also makes it possible to provide theoretical guarantees on the robustness of the model against a well-defined set of adversarial capabilities. Our empirical exploration validates our theoretical insights, where we can consistently reach 0% Attack Success Rate after a few adversarial retraining iterations.
- Open Access
- 1 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 383-398
Unmasking the Veiled: A Comprehensive Analysis of Android Evasive Malware
Since Android is the most widespread operating system, malware targeting it poses a severe threat to the security and privacy of millions of users and is increasing from year to year. The response from the community was swift, and many researchers have ventured to defend this system. In this cat-and-mouse game, attackers pay special attention to flying under the radar of analysis tools, and the techniques to understand whether their app is under analysis have become more and more sophisticated. Moreover, these evasive techniques are also adopted by benign apps to deter reverse engineering, making this phenomenon pervasive in the Android app ecosystem. While the scientific literature has proposed many evasive techniques and investigated their impact, one aspect still needs to be studied: how and to what extent Android apps, both malware and goodware, use such controls. This paper fills this gap by introducing a comprehensive taxonomy of evasive controls for the Android ecosystem and a proof-of-concept app that implements them all. We release the app as open source to help researchers and practitioners to assess whether their app analysis systems are sufficiently resilient to known evasion techniques. We also propose DroidDungeon, a novel probe-based sandbox, which circumvents evasive techniques thanks to a substantial engineering effort, making the apps under analysis believe they are running on an actual device. To the best of our knowledge, currently, DroidDungeon is the only solution providing anti-evasion capabilities, maintainability, and scalability at once. Using our sandbox, we studied evasive controls in both benign and malicious Android apps, revealing insights about their purpose, differences, and relationships between evasive controls and packers/protectors. Finally, we analyzed how the execution of an app differs depending on the presence or absence of evasive counter-measures. Our main finding is that 14% and 4% of malicious and benign samples refrain from running in an analysis environment that does not correctly mitigate evasive controls.
- 1 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 383-398
Open AccessUnmasking the Veiled: A Comprehensive Analysis of Android Evasive Malware
Since Android is the most widespread operating system, malware targeting it poses a severe threat to the security and privacy of millions of users and is increasing from year to year. The response from the community was swift, and many researchers have ventured to defend this system. In this cat-and-mouse game, attackers pay special attention to flying under the radar of analysis tools, and the techniques to understand whether their app is under analysis have become more and more sophisticated. Moreover, these evasive techniques are also adopted by benign apps to deter reverse engineering, making this phenomenon pervasive in the Android app ecosystem. While the scientific literature has proposed many evasive techniques and investigated their impact, one aspect still needs to be studied: how and to what extent Android apps, both malware and goodware, use such controls. This paper fills this gap by introducing a comprehensive taxonomy of evasive controls for the Android ecosystem and a proof-of-concept app that implements them all. We release the app as open source to help researchers and practitioners to assess whether their app analysis systems are sufficiently resilient to known evasion techniques. We also propose DroidDungeon, a novel probe-based sandbox, which circumvents evasive techniques thanks to a substantial engineering effort, making the apps under analysis believe they are running on an actual device. To the best of our knowledge, currently, DroidDungeon is the only solution providing anti-evasion capabilities, maintainability, and scalability at once. Using our sandbox, we studied evasive controls in both benign and malicious Android apps, revealing insights about their purpose, differences, and relationships between evasive controls and packers/protectors. Finally, we analyzed how the execution of an app differs depending on the presence or absence of evasive counter-measures. Our main finding is that 14% and 4% of malicious and benign samples refrain from running in an analysis environment that does not correctly mitigate evasive controls.
- Open Access
- 15 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 60-74
Decoding the Secrets of Machine Learning in Malware Classification: A Deep Dive into Datasets, Feature Extraction, and Model Performance
Many studies have proposed machine-learning (ML) models for malware detection and classification, reporting an almost-perfect performance. However, they assemble ground-truth in different ways, use diverse static- and dynamic-analysis techniques for feature extraction, and even differ on what they consider a malware family. As a consequence, our community still lacks an understanding of malware classification results: whether they are tied to the nature and distribution of the collected dataset, to what extent the number of families and samples in the training dataset influence performance, and how well static and dynamic features complement each other. This work sheds light on those open questions by investigating the impact of datasets, features, and classifiers on ML-based malware detection and classification. For this, we collect the largest balanced malware dataset so far with 67k samples from 670 families (100 samples each), and train state-of-the-art models for malware detection and family classification using our dataset. Our results reveal that static features perform better than dynamic features, and that combining both only provides marginal improvement over static features. We discover no correlation between packing and classification accuracy, and that missing behaviors in dynamically-extracted features highly penalise their performance. We also demonstrate how a larger number of families to classify makes the classification harder, while a higher number of samples per family increases accuracy. Finally, we find that models trained on a uniform distribution of samples per family better generalize on unseen data.
- 15 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 60-74
Open AccessDecoding the Secrets of Machine Learning in Malware Classification: A Deep Dive into Datasets, Feature Extraction, and Model Performance
Many studies have proposed machine-learning (ML) models for malware detection and classification, reporting an almost-perfect performance. However, they assemble ground-truth in different ways, use diverse static- and dynamic-analysis techniques for feature extraction, and even differ on what they consider a malware family. As a consequence, our community still lacks an understanding of malware classification results: whether they are tied to the nature and distribution of the collected dataset, to what extent the number of families and samples in the training dataset influence performance, and how well static and dynamic features complement each other. This work sheds light on those open questions by investigating the impact of datasets, features, and classifiers on ML-based malware detection and classification. For this, we collect the largest balanced malware dataset so far with 67k samples from 670 families (100 samples each), and train state-of-the-art models for malware detection and family classification using our dataset. Our results reveal that static features perform better than dynamic features, and that combining both only provides marginal improvement over static features. We discover no correlation between packing and classification accuracy, and that missing behaviors in dynamically-extracted features highly penalise their performance. We also demonstrate how a larger number of families to classify makes the classification harder, while a higher number of samples per family increases accuracy. Finally, we find that models trained on a uniform distribution of samples per family better generalize on unseen data.
- Open Access
- 5 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
The Dark Side of Native Code on Android
From a little research experiment to an essential component of military arsenals, malicious software has constantly been growing and evolving for more than three decades. On the other hand, from a negligible market share, the Android operating system is nowadays the most widely used mobile operating system, becoming a desirable target for large-scale malware distribution. While scientific literature has followed this trend, one aspect has been understudied: the role of native code in malicious Android apps. Android apps are written in high-level languages, but thanks to the Java Native Interface (JNI), Android also supports calling native (C/C++) library functions. While allowing native code in Android apps has a strong positive impact from a performance perspective, it dramatically complicates its analysis because bytecode and native code need different abstractions and analysis algorithms, and they thus pose different challenges and limitations. Consequently, these difficulties are often (ab)used to hide malicious payloads. In this work, we propose a novel methodology to reverse engineering Android apps focusing on suspicious patterns related to native components, i.e., surreptitious code that requires further inspection. We implemented a static analysis tool based on such methodology, which can bridge the “Java” and the native worlds and perform an in-depth analysis of tag code blocks responsible for suspicious behavior. These tags benefit the human facing the reverse engineering task: they clearly indicate which part of the code to focus on to find malicious code. Then, we performed a longitudinal analysis of Android malware over the past ten years and compared the recent malicious samples with actual top apps on the Google Play Store. Our work depicts typical behaviors of modern malware, its evolution, and how it abuses the native layer to complicate the analysis, especially with dynamic code loading and novel anti-analysis techniques. Finally, we show a use case for our suspicious tags: we trained and tested a machine learning algorithm for a binary classification task. Even if suspicious does not imply malicious, our classifier obtained a remarkable F1-score of 0.97, showing that our methodology can be helpful to both humans and machines.
- 5 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessThe Dark Side of Native Code on Android
From a little research experiment to an essential component of military arsenals, malicious software has constantly been growing and evolving for more than three decades. On the other hand, from a negligible market share, the Android operating system is nowadays the most widely used mobile operating system, becoming a desirable target for large-scale malware distribution. While scientific literature has followed this trend, one aspect has been understudied: the role of native code in malicious Android apps. Android apps are written in high-level languages, but thanks to the Java Native Interface (JNI), Android also supports calling native (C/C++) library functions. While allowing native code in Android apps has a strong positive impact from a performance perspective, it dramatically complicates its analysis because bytecode and native code need different abstractions and analysis algorithms, and they thus pose different challenges and limitations. Consequently, these difficulties are often (ab)used to hide malicious payloads. In this work, we propose a novel methodology to reverse engineering Android apps focusing on suspicious patterns related to native components, i.e., surreptitious code that requires further inspection. We implemented a static analysis tool based on such methodology, which can bridge the “Java” and the native worlds and perform an in-depth analysis of tag code blocks responsible for suspicious behavior. These tags benefit the human facing the reverse engineering task: they clearly indicate which part of the code to focus on to find malicious code. Then, we performed a longitudinal analysis of Android malware over the past ten years and compared the recent malicious samples with actual top apps on the Google Play Store. Our work depicts typical behaviors of modern malware, its evolution, and how it abuses the native layer to complicate the analysis, especially with dynamic code loading and novel anti-analysis techniques. Finally, we show a use case for our suspicious tags: we trained and tested a machine learning algorithm for a binary classification task. Even if suspicious does not imply malicious, our classifier obtained a remarkable F1-score of 0.97, showing that our methodology can be helpful to both humans and machines.