no code implementations • 19 Apr 2024 • Denis Donadel, Francesco Marchiori, Luca Pajola, Mauro Conti
Private LLMs achieve noteworthy results in small and medium networks, while challenges persist in comprehending complex network topologies, particularly for open-source models.
1 code implementation • 5 Mar 2024 • Ehsan Nowroozi, Imran Haider, Rahim Taheri, Mauro Conti
In LF, we randomly flipped the labels of benign data and trained the model on the manipulated data.
no code implementations • 25 Jan 2024 • Pier Paolo Tricomi, Luca Pajola, Luca Pasa, Mauro Conti
In this work, we investigate the relationship between Spotify users' attributes and their public playlists.
no code implementations • 7 Jan 2024 • Sameera K. M., Serena Nicolazzo, Marco Arazzi, Antonino Nocera, Rafidha Rehiman K. A., Vinod P, Mauro Conti
Federated Learning (FL) has recently arisen as a revolutionary approach to collaborative training Machine Learning models.
no code implementations • 8 Dec 2023 • Florine W. Dekker, Zekeriya Erkin, Mauro Conti
Specifically, we show that reconstruction requires a number of adversaries linear in the length of the network's shortest cycle.
no code implementations • 6 Dec 2023 • Matteo Gioele Collu, Tom Janssen-Groesbeek, Stefanos Koffas, Mauro Conti, Stjepan Picek
This work shows that by using adversarial personas, one can overcome safety mechanisms set out by ChatGPT and Bard.
no code implementations • 12 Oct 2023 • Mauro Conti, Nicola Farronato, Stefanos Koffas, Luca Pajola, Stjepan Picek
Optical Character Recognition (OCR) is a widely used tool to extract text from scanned documents.
1 code implementation • 4 Oct 2023 • Filippo Perrina, Francesco Marchiori, Mauro Conti, Nino Vincenzo Verde
In this paper, we introduce AGIR (Automatic Generation of Intelligence Reports), a transformative Natural Language Generation tool specifically designed to address the pressing challenges in the realm of CTI reporting.
2 code implementations • 7 Sep 2023 • Francesco Marchiori, Mauro Conti
By using our proposed methodologies, manufacturers can ensure that devices only use legitimate batteries, guaranteeing the operational state of any system and safety measures for the users.
no code implementations • 4 Aug 2023 • Marco Arazzi, Mauro Conti, Stefanos Koffas, Marina Krcek, Antonino Nocera, Stjepan Picek, Jing Xu
In this work, we are the first (to the best of our knowledge) to investigate label inference attacks on VFL using a zero-background knowledge strategy.
1 code implementation • 27 Jun 2023 • Marco Alecci, Mauro Conti, Francesco Marchiori, Luca Martinelli, Luca Pajola
An alarming side-effect of evasion attacks is their ability to transfer among different models: this property is called transferability.
no code implementations • 9 May 2023 • Marco Arazzi, Mauro Conti, Antonino Nocera, Stjepan Picek
Recently, researchers have successfully employed Graph Neural Networks (GNNs) to build enhanced recommender systems due to their capability to learn patterns from the interaction between involved entities.
1 code implementation • 27 Apr 2023 • Nicholas Boucher, Luca Pajola, Ilia Shumailov, Ross Anderson, Mauro Conti
Search engines are vulnerable to attacks against indexing and searching via text encoding manipulation.
no code implementations • 6 Apr 2023 • Ehsan Nowroozi, Yoosef Habibi, Mauro Conti
To highlight the problems involved with the evaluation of the dataset's IRIS images, we conducted a large number of analyses employing Siamese Neural Networks to assess the similarities between genuine and synthetic human IRISes, such as ResNet50, Xception, VGG16, and MobileNet-v2.
no code implementations • 31 Mar 2023 • Sara Bardi, Mauro Conti, Luca Pajola, Pier Paolo Tricomi
However, by choosing an appropriate content topic, this attractive mechanism could be extended to any OSN users, rather than only luring malicious actors.
no code implementations • 17 Mar 2023 • Francesco Marchiori, Mauro Conti, Nino Vincenzo Verde
The automatic extraction of information from Cyber Threat Intelligence (CTI) reports is crucial in risk management.
no code implementations • 6 Mar 2023 • Ehsan Nowroozi, Seyedsadra Seyedshoari, Yassine Mekdad, Erkay Savas, Mauro Conti
Digital wallet as a software program or a digital device allows users to conduct various transactions.
no code implementations • 3 Feb 2023 • Gorka Abad, Jing Xu, Stefanos Koffas, Behrad Tajalli, Stjepan Picek, Mauro Conti
Nevertheless, it is vulnerable to backdoor attacks that modify the training set to embed a secret functionality in the trained model.
no code implementations • 17 Jan 2023 • Serena Tardelli, Leonardo Nizzoli, Maurizio Tesconi, Mauro Conti, Preslav Nakov, Giovanni Da San Martino, Stefano Cresci
Large-scale online campaigns, malicious or otherwise, require a significant degree of coordination among participants, which sparked interest in the study of coordinated online behavior.
no code implementations • 17 Jan 2023 • Pier Paolo Tricomi, Marco Chilese, Mauro Conti, Ahmad-Reza Sadeghi
Thanks to our interpretable approaches, we conclude by outlining guidelines for creating successful posts.
1 code implementation • 6 Nov 2022 • Stefanos Koffas, Luca Pajola, Stjepan Picek, Mauro Conti
This work explores stylistic triggers for backdoor attacks in the audio domain: dynamic transformations of malicious samples through guitar effects.
no code implementations • 28 Oct 2022 • Mauro Conti, Jiaxin Li, Stjepan Picek
Membership Inference Attacks (MIAs) infer whether a data point is in the training data of a machine learning model.
1 code implementation • 24 Oct 2022 • Ying Yuan, Giovanni Apruzzese, Mauro Conti
By considering the application of ML for Phishing Website Detection (PWD), we formalize the "evasion-space" in which an adversarial perturbation can be introduced to fool a ML-PWD -- demonstrating that even perturbations in the "feature-space" are useful.
1 code implementation • 17 Oct 2022 • Pier Paolo Tricomi, Lisa Facciolo, Giovanni Apruzzese, Mauro Conti
This paper is the first to investigate such a problem.
no code implementations • 25 Sep 2022 • Ehsan Nowroozi, Mohammadreza Mohammadi, Erkay Savas, Mauro Conti, Yassine Mekdad
In this study, we present a novel architecture based on an ensemble classifier that combines the enhanced security of 1-Class classification (known as 1C) with the high performance of conventional 2-Class classification (known as 2C) in the absence of attacks. Our architecture is referred to as the 1. 5-Class (SPRITZ-1. 5C) classifier and constructed using a final dense classifier, one 2C classifier (i. e., CNNs), and two parallel 1C classifiers (i. e., auto-encoders).
1 code implementation • 11 Sep 2022 • Ehsan Nowroozi, Mohammadreza Mohammadi, Pargol Golmohammadi, Yassine Mekdad, Mauro Conti, Selcuk Uluagac
Our novel approach consists of changing the training strategy in the target network classifier and selecting random feature samples.
no code implementations • 27 Jul 2022 • Mauro Conti, Jiaxin Li, Stjepan Picek, Jing Xu
Even in those scenarios, our label-only MIA achieves a better attack performance in most cases.
1 code implementation • 27 Apr 2022 • Ehsan Nowroozi, abhishek, Mohammadreza Mohammadi, Mauro Conti
In this study, we extract a novel set of lexical and web-scrapped features and employ machine learning technique to set up system for fraudulent advertisement URLs detection.
no code implementations • 25 Apr 2022 • Ehsan Nowroozi, Yassine Mekdad, Mauro Conti, Simone Milani, Selcuk Uluagac, Berrin Yanikoglu
Additionally, it enables users to employ a virtual background to conceal their own environment due to privacy concerns or to reduce distractions, particularly in professional settings.
1 code implementation • 9 Mar 2022 • Giovanni Apruzzese, Luca Pajola, Mauro Conti
By using XeNIDS on six well-known datasets, we demonstrate the concealed potential, but also the risks, of cross-evaluations of ML-NIDS.
no code implementations • 4 Mar 2022 • Stefanos Koffas, Stjepan Picek, Mauro Conti
It was recently shown that countermeasures in image classification, like Neural Cleanse and ABS, could be bypassed with dynamic triggers that are effective regardless of their pattern and location.
no code implementations • 3 Mar 2022 • Ehsan Nowroozi, Mauro Conti, Yassine Mekdad
On the other hand, the recent development of GAN models may create high-quality face images without evidence of spatial artifacts.
no code implementations • 11 Jan 2022 • Mauro Conti, Luca Pajola, Pier Paolo Tricomi
Content moderators constantly monitor these online platforms to prevent the spreading of inappropriate content (e. g., hate speech, nudity images).
no code implementations • 15 Oct 2021 • Matteo Cardaioli, Stefano Cecconello, Mauro Conti, Simone Milani, Stjepan Picek, Eugen Saraci
We consider the setting where the attacker can access an ATM PIN pad of the same brand/model as the target one.
no code implementations • 9 Oct 2021 • Ehsan Nowroozi, Yassine Mekdad, Mohammad Hajian Berenjestanaki, Mauro Conti, Abdeslam El Fergougui
In this paper, we provide the first comprehensive study which assesses the robustness of CNN-based models for computer networks against adversarial transferability.
no code implementations • 27 Sep 2021 • Kristina Hristakieva, Stefano Cresci, Giovanni Da San Martino, Mauro Conti, Preslav Nakov
Large-scale manipulations on social media have two important characteristics: (i) use of propaganda to influence others, and (ii) adoption of coordinated behavior to spread it and to amplify its impact.
no code implementations • 30 Jul 2021 • Stefanos Koffas, Jing Xu, Mauro Conti, Stjepan Picek
This work explores backdoor attacks for automatic speech recognition systems where we inject inaudible triggers.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +2
no code implementations • 29 Jun 2021 • Mauro Conti, Simone Milani, Ehsan Nowroozi, Gabriele Orazi
On the other hand, users maybe want to fool people in the meeting by considering the virtual background to conceal where they are.
1 code implementation • 13 Apr 2021 • Luca Pajola, Mauro Conti
The increased demand for machine learning applications made companies offer Machine-Learning-as-a-Service (MLaaS).
no code implementations • 12 Feb 2021 • Francesco Betti Sorbelli, Mauro Conti, Cristina M. Pinotti, Giulio Rigoni
No specific attacks and defenses have been found in literature for GNSS+ or for UAVs moving in group without a pre-ordered arrangement.
Cryptography and Security
no code implementations • 10 Feb 2021 • Mauro Conti, Denis Donadel, Federico Turrin
In dealing with this security requirement, the research community focuses on developing new security mechanisms such as Intrusion Detection Systems (IDSs), facilitated by leveraging modern machine learning techniques.
Intrusion Detection Cryptography and Security
no code implementations • 28 Sep 2020 • Federico Tavella, Alberto Giaretta, Mauro Conti, Sasitharan Balasubramaniam
The similarities between these biological media and classical ones can also be a drawback, as malicious parties might replicate traditional attacks on the former archival system, using biological instruments and techniques.
no code implementations • 2 Jul 2020 • Giovanni Barbieri, Mauro Conti, Nils Ole Tippenhauer, Federico Turrin
Therefore, Shodan do not allow to understand the actual use of insecure industrial protocols on the Internet and the current security practices in ICS communications.
Cryptography and Security Networking and Internet Architecture
no code implementations • 13 Aug 2019 • Rahim Taheri, Meysam Ghahramani, Reza Javidan, Mohammad Shojafar, Zahra Pooranian, Mauro Conti
We test our experiments in a different type of features: API, intent, and permission features on these three datasets.
no code implementations • 13 Aug 2019 • Rahim Taheri, Reza Javidan, Mohammad Shojafar, Zahra Pooranian, Ali Miri, Mauro Conti
Our evaluation shows that using random forest feature selection and varying ratios of features can result in an improvement of up to 19\% accuracy when compared with the state-of-the-art method in the literature.
no code implementations • 20 Apr 2019 • Rahim Taheri, Reza Javidan, Mohammad Shojafar, Vinod P, Mauro Conti
We also test our methods using various classifier algorithms and compare them with the state-of-the-art data poisoning method using the Jacobian matrix.
no code implementations • 30 Mar 2019 • Kiran Balagani, Matteo Cardaioli, Mauro Conti, Paolo Gasti, Martin Georgiev, Tristan Gurtler, Daniele Lain, Charissa Miller, Kendall Molas, Nikita Samarin, Eugen Saraci, Gene Tsudik, Lynn Wu
This paper studies leakage of user passwords and PINs based on observations of typing feedback on screens or from projectors in the form of masked characters that indicate keystrokes.
Cryptography and Security K.6.5
no code implementations • 17 Sep 2018 • Deepa K, Radhamani G, Vinod P, Mohammad Shojafar, Neeraj Kumar, Mauro Conti
Ever increasing number of Android malware, has always been a concern for cybersecurity professionals.
no code implementations • 28 Aug 2018 • Tommi Gröndahl, Luca Pajola, Mika Juuti, Mauro Conti, N. Asokan
With the spread of social networks and their unfortunate use for hate speech, automatic detection of the latter has become a pressing problem.
no code implementations • 8 Aug 2018 • Abbas Acar, Hossein Fereidooni, Tigist Abera, Amit Kumar Sikder, Markus Miettinen, Hidayet Aksu, Mauro Conti, Ahmad-Reza Sadeghi, A. Selcuk Uluagac
It is realized utilizing state-of-the-art machine-learning approaches for detecting and identifying particular types of IoT devices, their actions, states, and ongoing user activities in a cascading style by only observing passively the wireless traffic from smart home devices.
Cryptography and Security
no code implementations • 27 Jul 2018 • Mauro Conti, Ali Dehghantanha, Katrin Franke, Steve Watson
The Internet of Things (IoT) envisions pervasive, connected, and smart nodes interacting autonomously while offering all sorts of services.
Cryptography and Security
no code implementations • 15 Sep 2017 • Tooska Dargahi, Ali Dehghantanha, Mauro Conti
Voice over Internet Protocol (VoIP) applications (apps) provide convenient and low cost means for users to communicate and share information with each other in real-time.
Cryptography and Security
1 code implementation • 28 Jun 2017 • Hossein Fereidooni, Jiska Classen, Tom Spink, Paul Patras, Markus Miettinen, Ahmad-Reza Sadeghi, Matthias Hollick, Mauro Conti
In this paper, we provide an in-depth security analysis of the operation of fitness trackers commercialized by Fitbit, the wearables market leader.
Cryptography and Security
no code implementations • 30 Nov 2016 • Parvez Faruki, Hossein Fereidooni, Vijay Laxmi, Mauro Conti, Manoj Gaur
We believe that, there is a need to investigate efficiency of the defense techniques used for code protection.
Cryptography and Security
no code implementations • 21 Oct 2016 • Moreno Ambrosin, Paolo Braca, Mauro Conti, Riccardo Lazzaretti
ODIN is a privacy-preserving extension of the popular consensus gossip algorithm, that prevents distributed agents have direct access to the data while they iteratively reach consensus; agents cannot access even the final consensus value, but can only retrieve partial information, e. g., a binary decision.
Cryptography and Security