Search

In a classification task, counterfactual explanations provide the minimum change needed for an input to be classified into a favorable class. We consider the problem of privately retrieving the exact closest counterfactual from a database of accepted samples while enforcing that certain features of the input sample cannot be changed, i.e., they are \emph{immutable}. An applicant (user) whose feature vector is rejected by a machine learning model wants to retrieve the sample closest to them in the database without altering a private subset of their features, which constitutes the immutable set. While doing this, the user should keep their feature vector, immutable set and the resulting counterfactual index information-theoretically private from the institution. We refer to this as immutable private counterfactual retrieval (I-PCR) problem which generalizes PCR to a more practical setting. In this paper, we propose two I-PCR schemes by leveraging techniques from private information retrieval (PIR) and characterize their communication costs. Further, we quantify the information that the user learns about the database and compare it for the proposed schemes.

Automated marker makers (AMMs) are a class of decentralized exchanges that enable the automated trading of digital assets. They accept deposits of digital tokens from liquidity providers (LPs); tokens can be used by traders to execute trades, which generate fees for the investing LPs. The distinguishing feature of AMMs is that trade prices are determined algorithmically, unlike classical limit order books. Concentrated liquidity market makers (CLMMs) are a major class of AMMs that offer liquidity providers flexibility to decide not only \emph{how much} liquidity to provide, but \emph{in what ranges of prices} they want the liquidity to be used. This flexibility can complicate strategic planning, since fee rewards are shared among LPs. We formulate and analyze a game theoretic model to study the incentives of LPs in CLMMs. Our main results show that while our original formulation admits multiple Nash equilibria and has complexity quadratic in the number of price ticks in the contract, it can be reduced to a game with a unique Nash equilibrium whose complexity is only linear. We further show that the Nash equilibrium of this simplified game follows a waterfilling strategy, in which low-budget LPs use up their full budget, but rich LPs do not. Finally, by fitting our game model to real-world CLMMs, we observe that in liquidity pools with risky assets, LPs adopt investment strategies far from the Nash equilibrium. Under price uncertainty, they generally invest in fewer and wider price ranges than our analysis suggests, with lower-frequency liquidity updates. We show that across several pools, by updating their strategy to more closely match the Nash equilibrium of our game, LPs can improve their median daily returns by \$116, which corresponds to an increase of 0.009\% in median daily return on investment.

In recent years, text-to-image (T2I) generation models have made significant progress in generating high-quality images that align with text descriptions. However, these models also face the risk of unsafe generation, potentially producing harmful content that violates usage policies, such as explicit material. Existing safe generation methods typically focus on suppressing inappropriate content by erasing undesired concepts from visual representations, while neglecting to sanitize the textual representation. Although these methods help mitigate the risk of misuse to certain extent, their robustness remains insufficient when dealing with adversarial attacks. Given that semantic consistency between input text and output image is a fundamental requirement for T2I models, we identify that textual representations (i.e., prompt embeddings) are likely the primary source of unsafe generation. To this end, we propose a vision-agnostic safe generation framework, Embedding Sanitizer (ES), which focuses on erasing inappropriate concepts from prompt embeddings and uses the sanitized embeddings to guide the model for safe generation. ES is applied to the output of the text encoder as a plug-and-play module, enabling seamless integration with different T2I models as well as other safeguards. In addition, ES's unique scoring mechanism assigns a score to each token in the prompt to indicate its potential harmfulness, and dynamically adjusts the sanitization intensity to balance defensive performance and generation quality. Through extensive evaluation on five prompt benchmarks, our approach achieves state-of-the-art robustness by sanitizing the source (prompt embedding) of unsafe generation compared to nine baseline methods. It significantly outperforms existing safeguards in terms of interpretability and controllability while maintaining generation quality.

Modern cryptography, such as Rivest Shamir Adleman (RSA) and Secure Hash Algorithm (SHA), has been designed by humans based on our understanding of cryptographic methods. Neural Network (NN) based cryptography is being investigated due to its ability to learn and implement random cryptographic schemes that may be harder to decipher than human-designed algorithms. NN based cryptography may create a new cryptographic scheme that is NN specific and that changes every time the NN is (re)trained. This is attractive since it would require an adversary to restart its process(es) to learn or break the cryptographic scheme every time the NN is (re)trained. Current challenges facing NN-based encryption include additional communication overhead due to encoding to correct bit errors, quantizing the continuous-valued output of the NN, and enabling One-Time-Pad encryption. With this in mind, the Random Adversarial Data Obfuscation Model (RANDOM) Adversarial Neural Cryptography (ANC) network is introduced. RANDOM is comprised of three new NN layers: the (i) projection layer, (ii) inverse projection layer, and (iii) dot-product layer. This results in an ANC network that (i) is computationally efficient, (ii) ensures the encrypted message is unique to the encryption key, and (iii) does not induce any communication overhead. RANDOM only requires around 100 KB to store and can provide up to 2.5 megabytes per second of end-to-end encrypted communication., Comment: 8 pages

Lateral movement is a crucial component of advanced persistent threat (APT) attacks in networks. Attackers exploit security vulnerabilities in internal networks or IoT devices, expanding their control after initial infiltration to steal sensitive data or carry out other malicious activities, posing a serious threat to system security. Existing research suggests that attackers generally employ seemingly unrelated operations to mask their malicious intentions, thereby evading existing lateral movement detection methods and hiding their intrusion traces. In this regard, we analyze host authentication log data from a graph perspective and propose a multi-scale lateral movement detection framework called LMDetect. The main workflow of this framework proceeds as follows: 1) Construct a heterogeneous multigraph from host authentication log data to strengthen the correlations among internal system entities; 2) Design a time-aware subgraph generator to extract subgraphs centered on authentication events from the heterogeneous authentication multigraph; 3) Design a multi-scale attention encoder that leverages both local and global attention to capture hidden anomalous behavior patterns in the authentication subgraphs, thereby achieving lateral movement detection. Extensive experiments on two real-world authentication log datasets demonstrate the effectiveness and superiority of our framework in detecting lateral movement behaviors.

The integration of intelligent and connected technologies in modern vehicles, while offering enhanced functionalities through Electronic Control Unit and interfaces like OBD-II and telematics, also exposes the vehicle's in-vehicle network (IVN) to potential cyberattacks. In this paper, we consider a specific type of cyberattack known as the injection attack. As demonstrated by empirical data from real-world cybersecurity adversarial competitions(available at https://mimic2024.xctf.org.cn/race/qwmimic2024 ), these injection attacks have excitation effect over time, gradually manipulating network traffic and disrupting the vehicle's normal functioning, ultimately compromising both its stability and safety. To profile the abnormal behavior of attackers, we propose a novel injection attack detector to extract long-term features of attack behavior. Specifically, we first provide a theoretical analysis of modeling the time-excitation effects of the attack using Multi-Dimensional Hawkes Process (MDHP). A gradient descent solver specifically tailored for MDHP, MDHP-GDS, is developed to accurately estimate optimal MDHP parameters. We then propose an injection attack detector, MDHP-Net, which integrates optimal MDHP parameters with MDHP-LSTM blocks to enhance temporal feature extraction. By introducing MDHP parameters, MDHP-Net captures complex temporal features that standard Long Short-Term Memory (LSTM) cannot, enriching temporal dependencies within our customized structure. Extensive evaluations demonstrate the effectiveness of our proposed detection approach.

The high complexity of DNS poses unique challenges for ensuring its security and reliability. Despite continuous advances in DNS testing, monitoring, and verification, protocol-level defects still give rise to numerous bugs and attacks. In this paper, we provide the first decision procedure for the DNS verification problem, establishing its complexity as $\mathsf{2ExpTime}$, which was previously unknown. We begin by formalizing the semantics of DNS as a system of recursive communicating processes extended with timers and an infinite message alphabet. We provide an algebraic abstraction of the alphabet with finitely many equivalence classes, using the subclass of semigroups that recognize positive prefix-testable languages. We then introduce a novel generalization of bisimulation for labelled transition systems, weaker than strong bisimulation, to show that our abstraction is sound and complete. Finally, using this abstraction, we reduce the DNS verification problem to the verification problem for pushdown systems. To show the expressiveness of our framework, we model two of the most prominent attack vectors on DNS, namely amplification attacks and rewrite blackholing., Comment: Proceedings of the ACM on Programming Languages (POPL) 2025

During the past decade, Deep Neural Networks (DNNs) proved their value on a large variety of subjects. However despite their high value and public accessibility, the protection of the intellectual property of DNNs is still an issue and an emerging research field. Recent works have successfully extracted fully-connected DNNs using cryptanalytic methods in hard-label settings, proving that it was possible to copy a DNN with high fidelity, i.e., high similitude in the output predictions. However, the current cryptanalytic attacks cannot target complex, i.e., not fully connected, DNNs and are limited to special cases of neurons present in deep networks. In this work, we introduce a new end-to-end attack framework designed for model extraction of embedded DNNs with high fidelity. We describe a new black-box side-channel attack which splits the DNN in several linear parts for which we can perform cryptanalytic extraction and retrieve the weights in hard-label settings. With this method, we are able to adapt cryptanalytic extraction, for the first time, to non-fully connected DNNs, while maintaining a high fidelity. We validate our contributions by targeting several architectures implemented on a microcontroller unit, including a Multi-Layer Perceptron (MLP) of 1.7 million parameters and a shortened MobileNetv1. Our framework successfully extracts all of these DNNs with high fidelity (88.4% for the MobileNetv1 and 93.2% for the MLP). Furthermore, we use the stolen model to generate adversarial examples and achieve close to white-box performance on the victim's model (95.8% and 96.7% transfer rate).

The evolution of the Web3 ecosystem has been hindered by fragmented liquidity and limited interoperability across Layer 1 (L1) and Layer 2 (L2) blockchains, which leads to inefficiencies and elevated costs. Omnichain Web addresses these challenges by introducing a comprehensive framework to unify decentralized networks through its core components: OmniRollups, Proof Network, Ragno Network, and Builder Marketplace. This ecosystem enables seamless cross-chain asset settlement, interoperability, and user-friendly decentralized application (dApp) development, driven by innovative technologies such as modular proof networks and trusted execution environments (TEEs). By integrating advanced zero-knowledge proof systems and compatibility with AI agents, Omnichain Web empowers intent-driven and autonomous functionalities, streamlining liquidity management and user interactions across blockchains. Furthermore, its decentralized marketplace for L1 infrastructure reduces operational overhead and promotes scalable, secure, and efficient cross-chain protocols. As a pioneering solution, Omnichain Web seamlessly connects Web2 and Web3, enabling a holistic and interconnected digital economy.

Quantitative Information Flow (QIF) provides a robust information-theoretical framework for designing secure systems with minimal information leakage. While previous research has addressed the design of such systems under hard constraints (e.g. application limitations) and soft constraints (e.g. utility), scenarios often arise where the core system's behavior is considered fixed. In such cases, the challenge is to design a new component for the existing system that minimizes leakage without altering the original system. In this work we address this problem by proposing optimal solutions for constructing a new row, in a known and unmodifiable information-theoretic channel, aiming at minimizing the leakage. We first model two types of adversaries: an exact-guessing adversary, aiming to guess the secret in one try, and a s-distinguishing one, which tries to distinguish the secret s from all the other secrets.Then, we discuss design strategies for both fixed and unknown priors by offering, for each adversary, an optimal solution under linear constraints, using Linear Programming.We apply our approach to the problem of website fingerprinting defense, considering a scenario where a site administrator can modify their own site but not others. We experimentally evaluate our proposed solutions against other natural approaches. First, we sample real-world news websites and then, for both adversaries, we demonstrate that the proposed solutions are effective in achieving the least leakage. Finally, we simulate an actual attack by training an ML classifier for the s-distinguishing adversary and show that our approach decreases the accuracy of the attacker., Comment: 38th IEEE Computer Security Foundations Symposium, IEEE, Jun 2025, Santa Cruz, United States

Vibrometry-based side channels pose a significant privacy risk, exploiting sensors like mmWave radars, light sensors, and accelerometers to detect vibrations from sound sources or proximate objects, enabling speech eavesdropping. Despite various proposed defenses, these involve costly hardware solutions with inherent physical limitations. This paper presents EveGuard, a software-driven defense framework that creates adversarial audio, protecting voice privacy from side channels without compromising human perception. We leverage the distinct sensing capabilities of side channels and traditional microphones where side channels capture vibrations and microphones record changes in air pressure, resulting in different frequency responses. EveGuard first proposes a perturbation generator model (PGM) that effectively suppresses sensor-based eavesdropping while maintaining high audio quality. Second, to enable end-to-end training of PGM, we introduce a new domain translation task called Eve-GAN for inferring an eavesdropped signal from a given audio. We further apply few-shot learning to mitigate the data collection overhead for Eve-GAN training. Our extensive experiments show that EveGuard achieves a protection rate of more than 97 percent from audio classifiers and significantly hinders eavesdropped audio reconstruction. We further validate the performance of EveGuard across three adaptive attack mechanisms. We have conducted a user study to verify the perceptual quality of our perturbed audio.

As quantum computing continues to advance, its ability to compromise widely used cryptographic systems projects a significant challenge to modern cybersecurity. This paper outlines a strategic roadmap for industries to anticipate and mitigate the risks posed by quantum attacks. Our study explores the development of a quantum-resistant cryptographic solutioning framework for the industry, offering a practical and strategic approach to mitigating quantum attacks. We, here, propose a novel strategic framework, coined name STL-QCRYPTO, outlines tailored, industry-specific methodologies to implement quantum-safe security systems, ensuring long-term protection against the disruptive potential of quantum computing. The following fourteen high-risk sectors: Financial Services, Banking, Healthcare, Critical Infrastructure, Government & Defence, E-commerce, Energy & Utilities, Automotive & Transportation, Cloud Computing & Data Storage, Insurance, Internet & Telecommunications, Blockchain Applications, Metaverse Applications, and Multiagent AI Systems - are critically assessed for their vulnerability to quantum threats. The evaluation emphasizes practical approaches for the deployment of quantum-safe security systems to safeguard these industries against emerging quantum-enabled cyber risks. Additionally, the paper addresses the technical, operational, and regulatory hurdles associated with adopting quantum-resistant technologies. By presenting a structured timeline and actionable recommendations, this roadmap with proposed framework prepares industries with the essential strategy to safeguard their potential security threats in the quantum computing era.

Phishing has been a prevalent cyber threat that manipulates users into revealing sensitive private information through deceptive tactics, designed to masquerade as trustworthy entities. Over the years, proactively detection of phishing URLs (or websites) has been established as an widely-accepted defense approach. In literature, we often find supervised Machine Learning (ML) models with highly competitive performance for detecting phishing websites based on the extracted features from both phishing and benign (i.e., legitimate) websites. However, it is still unclear if these features or indicators are dependent on a particular dataset or they are generalized for overall phishing detection. In this paper, we delve deeper into this issue by analyzing two publicly available phishing URL datasets, where each dataset has its own set of unique and overlapping features related to URL string and website contents. We want to investigate if overlapping features are similar in nature across datasets and how does the model perform when trained on one dataset and tested on the other. We conduct practical experiments and leverage explainable AI (XAI) methods such as SHAP plots to provide insights into different features' contributions in case of phishing detection to answer our primary question, ``Can features for phishing URL detection be trusted across diverse dataset?''. Our case study experiment results show that features for phishing URL detection can often be dataset-dependent and thus may not be trusted across different datasets even though they share same set of feature behaviors., Comment: 8 pages, 10 figures, The 11th International Conference on Networking, Systems and Security, December 19-21, 2024

Machine learning (ML) defenses protect against various risks to security, privacy, and fairness. Real-life models need simultaneous protection against multiple different risks which necessitates combining multiple defenses. But combining defenses with conflicting interactions in an ML model can be ineffective, incurring a significant drop in the effectiveness of one or more defenses being combined. Practitioners need a way to determine if a given combination can be effective. Experimentally identifying effective combinations can be time-consuming and expensive, particularly when multiple defenses need to be combined. We need an inexpensive, easy-to-use combination technique to identify effective combinations. Ideally, a combination technique should be (a) accurate (correctly identifies whether a combination is effective or not), (b) scalable (allows combining multiple defenses), (c) non-invasive (requires no change to the defenses being combined), and (d) general (is applicable to different types of defenses). Prior works have identified several ad-hoc techniques but none satisfy all the requirements above. We propose a principled combination technique, Def\Con, to identify effective defense combinations. Def\Con meets all requirements, achieving 90% accuracy on eight combinations explored in prior work and 81% in 30 previously unexplored combinations that we empirically evaluate in this paper.

The use of Large Language Models (LLMs) in software development is rapidly growing, with developers increasingly relying on these models for coding assistance, including security-critical tasks. Our work presents a comprehensive comparison between traditional static analysis tools for cryptographic API misuse detection-CryptoGuard, CogniCrypt, and Snyk Code-and the LLMs-GPT and Gemini. Using benchmark datasets (OWASP, CryptoAPI, and MASC), we evaluate the effectiveness of each tool in identifying cryptographic misuses. Our findings show that GPT 4-o-mini surpasses current state-of-the-art static analysis tools on the CryptoAPI and MASC datasets, though it lags on the OWASP dataset. Additionally, we assess the quality of LLM responses to determine which models provide actionable and accurate advice, giving developers insights into their practical utility for secure coding. This study highlights the comparative strengths and limitations of static analysis versus LLM-driven approaches, offering valuable insights into the evolving role of AI in advancing software security practices.

The adoption of security protocols such as Transport Layer Security (TLS) has significantly improved the state of traffic encryption and integrity protection on the Internet. Despite rigorous analysis, vulnerabilities continue to emerge, sometimes due to fundamental flaws in the protocol specification. This paper examines the security of TLS when using Raw Public Key (RPK) authentication. This mode has not been as extensively studied as X.509 certificates and Pre-Shared Keys (PSK). We develop a formal model of TLS RPK using applied pi calculus and the ProVerif verification tool, revealing that the RPK mode is susceptible to identity misbinding attacks. Our contributions include formal models of TLS RPK with several mechanisms for binding the endpoint identity to its public key, verification results, practical scenarios demonstrating the misbinding attack, and recommendations for mitigating such vulnerabilities. These findings highlight the need for improved security measures in TLS RPK.

Differentiable rendering methods have emerged as a promising means for generating photo-realistic and physically plausible adversarial attacks by manipulating 3D objects and scenes that can deceive deep neural networks (DNNs). Recently, differentiable rendering capabilities have evolved significantly into a diverse landscape of libraries, such as Mitsuba, PyTorch3D, and methods like Neural Radiance Fields and 3D Gaussian Splatting for solving inverse rendering problems that share conceptually similar properties commonly used to attack DNNs, such as back-propagation and optimization. However, the adversarial machine learning research community has not yet fully explored or understood such capabilities for generating attacks. Some key reasons are that researchers often have different attack goals, such as misclassification or misdetection, and use different tasks to accomplish these goals by manipulating different representation in a scene, such as the mesh or texture of an object. This survey adopts a task-oriented unifying framework that systematically summarizes common tasks, such as manipulating textures, altering illumination, and modifying 3D meshes to exploit vulnerabilities in DNNs. Our framework enables easy comparison of existing works, reveals research gaps and spotlights exciting future research directions in this rapidly evolving field. Through focusing on how these tasks enable attacks on various DNNs such as image classification, facial recognition, object detection, optical flow and depth estimation, our survey helps researchers and practitioners better understand the vulnerabilities of computer vision systems against photorealistic adversarial attacks that could threaten real-world applications.

Backdoor attacks undermine the integrity of machine learning models by allowing attackers to manipulate predictions using poisoned training data. Such attacks lead to targeted misclassification when specific triggers are present, while the model behaves normally under other conditions. This paper considers a post-training backdoor defense task, aiming to detoxify the backdoors in pre-trained models. We begin by analyzing the underlying issues of vanilla fine-tuning and observe that it is often trapped in regions with low loss for both clean and poisoned samples. Motivated by such observations, we propose Distance-Driven Detoxification (D3), an innovative approach that reformulates backdoor defense as a constrained optimization problem. Specifically, D3 promotes the model's departure from the vicinity of its initial weights, effectively reducing the influence of backdoors. Extensive experiments on state-of-the-art (SOTA) backdoor attacks across various model architectures and datasets demonstrate that D3 not only matches but often surpasses the performance of existing SOTA post-training defense techniques., Comment: Preprint version

We study the differentially private top-$k$ selection problem, aiming to identify a sequence of $k$ items with approximately the highest scores from $d$ items. Recent work by Gillenwater et al. (ICML '22) employs a direct sampling approach from the vast collection of $d^{\,\Theta(k)}$ possible length-$k$ sequences, showing superior empirical accuracy compared to previous pure or approximate differentially private methods. Their algorithm has a time and space complexity of $\tilde{O}(dk)$. In this paper, we present an improved algorithm with time and space complexity $O(d + k^2 / \epsilon \cdot \ln d)$, where $\epsilon$ denotes the privacy parameter. Experimental results show that our algorithm runs orders of magnitude faster than their approach, while achieving similar empirical accuracy., Comment: NeurIPS 2024

Embedding-as-a-Service (EaaS) has emerged as a successful business pattern but faces significant challenges related to various forms of copyright infringement, including API misuse and different attacks. Various studies have proposed backdoor-based watermarking schemes to protect the copyright of EaaS services. In this paper, we reveal that previous watermarking schemes possess semantic-independent characteristics and propose the Semantic Perturbation Attack (SPA). Our theoretical and experimental analyses demonstrate that this semantic-independent nature makes current watermarking schemes vulnerable to adaptive attacks that exploit semantic perturbations test to bypass watermark verification. To address this vulnerability, we propose the Semantic Aware Watermarking (SAW) scheme, a robust defense mechanism designed to resist SPA, by injecting a watermark that adapts to the text semantics. Extensive experimental results across multiple datasets demonstrate that the True Positive Rate (TPR) for detecting watermarked samples under SPA can reach up to more than 95%, rendering previous watermarks ineffective. Meanwhile, our watermarking scheme can resist such attack while ensuring the watermark verification capability. Our code is available at https://github.com/Zk4-ps/EaaS-Embedding-Watermark.

Trusted Execution Environments (TEE) are used to safeguard on-device models. However, directly employing TEEs to secure the entire DNN model is challenging due to the limited computational speed. Utilizing GPU can accelerate DNN's computation speed but commercial widely-available GPUs usually lack security protection. To this end, scholars introduce TSDP, a method that protects privacy-sensitive weights within TEEs and offloads insensitive weights to GPUs. Nevertheless, current methods do not consider the presence of a knowledgeable adversary who can access abundant publicly available pre-trained models and datasets. This paper investigates the security of existing methods against such a knowledgeable adversary and reveals their inability to fulfill their security promises. Consequently, we introduce a novel partition before training strategy, which effectively separates privacy-sensitive weights from other components of the model. Our evaluation demonstrates that our approach can offer full model protection with a computational cost reduced by a factor of 10. In addition to traditional CNN models, we also demonstrate the scalability to large language models. Our approach can compress the private functionalities of the large language model to lightweight slices and achieve the same level of protection as the shielding-whole-model baseline., Comment: Accepted by TOSEM. Extended version of the S&P24 paper (arXiv:2310.07152)

Virtual reality (VR), while enhancing user experiences, introduces significant privacy risks. This paper reveals a novel vulnerability in VR systems that allows attackers to capture VR privacy through obstacles utilizing millimeter-wave (mmWave) signals without physical intrusion and virtual connection with the VR devices. We propose mmSpyVR, a novel attack on VR user's privacy via mmWave radar. The mmSpyVR framework encompasses two main parts: (i) A transfer learning-based feature extraction model to achieve VR feature extraction from mmWave signal. (ii) An attention-based VR privacy spying module to spy VR privacy information from the extracted feature. The mmSpyVR demonstrates the capability to extract critical VR privacy from the mmWave signals that have penetrated through obstacles. We evaluate mmSpyVR through IRB-approved user studies. Across 22 participants engaged in four experimental scenes utilizing VR devices from three different manufacturers, our system achieves an application recognition accuracy of 98.5\% and keystroke recognition accuracy of 92.6\%. This newly discovered vulnerability has implications across various domains, such as cybersecurity, privacy protection, and VR technology development. We also engage with VR manufacturer Meta to discuss and explore potential mitigation strategies. Data and code are publicly available for scrutiny and research at https://github.com/luoyumei1-a/mmSpyVR/

To ensure secure and reliable communication in wireless systems, authenticating the identities of numerous nodes is imperative. Traditional cryptography-based authentication methods suffer from issues such as low compatibility, reliability, and high complexity. Physical-Layer Authentication (PLA) is emerging as a promising complement due to its exploitation of unique properties in wireless environments. Recently, Machine Learning (ML)-based PLA has gained attention for its intelligence, adaptability, universality, and scalability compared to non-ML approaches. However, a comprehensive overview of state-of-the-art ML-based PLA and its foundational aspects is lacking. This paper presents a comprehensive survey of characteristics and technologies that can be used in the ML-based PLA. We categorize existing ML-based PLA schemes into two main types: multi-device identification and attack detection schemes. In deep learning-based multi-device identification schemes, Deep Neural Networks are employed to train models, avoiding complex processing and expert feature transformation. Deep learning-based multi-device identification schemes are further subdivided, with schemes based on Convolutional Neural Networks being extensively researched. In ML-based attack detection schemes, receivers utilize intelligent ML techniques to set detection thresholds automatically, eliminating the need for manual calculation or knowledge of channel models. ML-based attack detection schemes are categorized into three sub-types: Supervised Learning, Unsupervised Learning, and Reinforcement Learning. Additionally, we summarize open-source datasets used for PLA, encompassing Radio Frequency fingerprints and channel fingerprints. Finally, this paper outlines future research directions to guide researchers in related fields., Comment: 111 pages, 9 figures

With the increasing emphasis on privacy regulations, such as GDPR, protecting individual privacy and ensuring compliance have become critical concerns for both individuals and organizations. Privacy-preserving machine learning (PPML) is an innovative approach that allows for secure data analysis while safeguarding sensitive information. It enables organizations to extract valuable insights from data without compromising privacy. Secure multi-party computation (MPC) is a key tool in PPML, as it allows multiple parties to jointly compute functions without revealing their private inputs, making it essential in multi-server environments. We address the performance overhead of existing maliciously secure protocols, particularly in finite rings like $\mathbb{Z}_{2^\ell}$, by introducing an efficient protocol for secure linear function evaluation. We implement our maliciously secure MPC protocol on GPUs, significantly improving its efficiency and scalability. We extend the protocol to handle linear and non-linear layers, ensuring compatibility with a wide range of machine-learning models. Finally, we comprehensively evaluate machine learning models by integrating our protocol into the workflow, enabling secure and efficient inference across simple and complex models, such as convolutional neural networks (CNNs).

After more than 40 years of development, the fundamental TCP/IP protocol suite, serving as the backbone of the Internet, is widely recognized for having achieved an elevated level of robustness and security. Distinctively, we take a new perspective to investigate the security implications of cross-layer interactions within the TCP/IP protocol suite caused by ICMP error messages. Through a comprehensive analysis of interactions among Wi-Fi, IP, ICMP, UDP, and TCP due to ICMP errors, we uncover several significant vulnerabilities, including information leakage, desynchronization, semantic gaps, and identity spoofing. These vulnerabilities can be exploited by off-path attackers to manipulate network traffic stealthily, affecting over 20% of popular websites and more than 89% of public Wi-Fi networks, thus posing risks to the Internet. By responsibly disclosing these vulnerabilities to affected vendors and proposing effective countermeasures, we enhance the robustness of the TCP/IP protocol suite, receiving acknowledgments from well-known organizations such as the Linux community, the OpenWrt community, the FreeBSD community, Wi-Fi Alliance, Qualcomm, HUAWEI, China Telecom, Alibaba, and H3C., Comment: 9 pages, 11 figures

The widespread use of image acquisition technologies, along with advances in facial recognition, has raised serious privacy concerns. Face de-identification usually refers to the process of concealing or replacing personal identifiers, which is regarded as an effective means to protect the privacy of facial images. A significant number of methods for face de-identification have been proposed in recent years. In this survey, we provide a comprehensive review of state-of-the-art face de-identification methods, categorized into three levels: pixel-level, representation-level, and semantic-level techniques. We systematically evaluate these methods based on two key criteria, the effectiveness of privacy protection and preservation of image utility, highlighting their advantages and limitations. Our analysis includes qualitative and quantitative comparisons of the main algorithms, demonstrating that deep learning-based approaches, particularly those using Generative Adversarial Networks (GANs) and diffusion models, have achieved significant advancements in balancing privacy and utility. Experimental results reveal that while recent methods demonstrate strong privacy protection, trade-offs remain in visual fidelity and computational complexity. This survey not only summarizes the current landscape but also identifies key challenges and future research directions in face de-identification.

Multimedia streaming protocols are becoming increasingly popular in Crowdsensing due to their ability to deliver high-quality video content over the internet in real-time. Streaming multimedia content, as in the context of live video streaming, requires high bandwidth and large storage capacity to ensure a sufficient throughput. Crowdsensing can distribute information about shared video contents among multiple users in network, reducing storage capacity and computational and bandwidth requirements. However, Crowdsensing introduces several security constraints that must be taken into account to ensure the confidentiality, integrity, and availability of the data. In the specific case of video streaming, commonly named as visual crowdsensing (VCS) within this context, data is transmitted over wireless networks, making it vulnerable to security breaches and susceptible to eavesdropping and interception by attackers. Multimedias often contains sensitive user data and may be subject to various privacy laws, including data protection laws and laws related to photography and video recording, based on local GDPR (General Data Protection Regulation). For this reason the realization of a secure protocol optimized for a distributed data streaming in real-time becomes increasingly important in crowdsensing and smart-enviroment context. In this article, we will discuss the use of a symmetric AES-CTR encryption based protocol for securing data streaming over a crowd-sensed network.

Improving cybersecurity education has become a priority for many countries and organizations worldwide. Computing societies and professional associations have recognized cybersecurity as a distinctive computing discipline and created specialized cybersecurity curricular guidelines. Higher education institutions are introducing new cybersecurity programs, attracting students to this expanding field. In this paper, we examined 101 study programs across 24 countries. Based on their analysis, we argue that top-ranked universities have not yet fully implemented the guidelines and offer programs that have "cyber" in their name but lack some essential elements of a cybersecurity program. In particular, most programs do not sufficiently cover non-technical components, such as law, policies, or risk management. Also, most programs teach knowledge and skills but do not expose students to experiential learning outside the traditional classroom (such as internships) to develop their competencies. As a result, graduates of these programs may not meet employer expectations and may require additional training. To help program directors and educators improve their programs and courses, this paper offers examples of effective practices from cybersecurity programs around the world and our teaching practice., Comment: Published in ACM SIGCSE 2025 conference proceedings, see https://doi.org/10.1145/3641554.3701976

To meet the diverse needs of users, the rapid advancement of cloud-edge-device collaboration has become a standard practice. However, this complex environment, particularly in untrusted (non-collaborative) scenarios, presents numerous security challenges. Authentication acts as the first line of defense and is fundamental to addressing these issues. Although many authentication and key agreement schemes exist, they often face limitations, such as being tailored to overly specific scenarios where devices authenticate solely with either the edge or the cloud, or being unsuitable for resource-constrained devices. To address these challenges, we propose an adaptive and efficient authentication and key agreement scheme (AEAKA) for Cloud-Edge-Device IoT environments. This scheme is highly adaptive and scalable, capable of automatically and dynamically initiating different authentication methods based on device requirements. Additionally, it employs an edge-assisted authentication approach to reduce the load on third-party trust authorities. Furthermore, we introduce a hash-based algorithm for the authentication protocol, ensuring a lightweight method suitable for a wide range of resource-constrained devices while maintaining security. AEAKA ensures that entities use associated authentication credentials, enhancing the privacy of the authentication process. Security proofs and performance analyses demonstrate that AEAKA outperforms other methods in terms of security and authentication efficiency., Comment: 17 pages,14 figures,submitted to Transactions on Dependable and Secure Computing in 30-May-2024

The growing complexity of Internet of Things (IoT) environments, particularly in cross-domain data sharing, presents significant security challenges. Existing data-sharing schemes often rely on computationally expensive cryptographic operations and centralized key management, limiting their effectiveness for resource-constrained devices. To address these issues, we propose an efficient, secure blockchain-based data-sharing scheme. First, our scheme adopts a distributed key generation method, which avoids single point of failure. This method also allows independent pseudonym generation and key updates, enhancing authentication flexibility while reducing computational overhead. Additionally, the scheme provides a complete data-sharing process, covering data uploading, storage, and sharing, while ensuring data traceability, integrity, and privacy. Security analysis shows that the proposed scheme is theoretically secure and resistant to various attacks, while performance evaluations demonstrate lower computational and communication overhead compared to existing solutions, making it both secure and efficient for IoT applications., Comment: 15 pages,10 figures, submitted to Transactions on Information Forensics & Security in 19-Sep-2024

We explore an emerging threat model for end-to-end (E2E) encrypted applications: an adversary sends chosen messages to a target client, thereby "injecting" adversarial content into the application state. Such state is subsequently encrypted and synchronized to an adversarially-visible storage. By observing the lengths of the resulting cloud-stored ciphertexts, the attacker backs out confidential information. We investigate this injection threat model in the context of state-of-the-art encrypted messaging applications that support E2E encrypted backups. We show proof-of-concept attacks that can recover information about E2E encrypted messages or attachments sent via WhatsApp, assuming the ability to compromise the target user's Google or Apple account (which gives access to encrypted backups). We also show weaknesses in Signal's encrypted backup design that would allow injection attacks to infer metadata including a target user's number of contacts and conversations, should the adversary somehow obtain access to the user's encrypted Signal backup. While we do not believe our results should be of immediate concern for users of these messaging applications, our results do suggest that more work is needed to build tools that enjoy strong E2E security guarantees., Comment: Published in IEEE Security and Privacy 2024

Smart contracts are software programs that enable diverse business activities on the blockchain. Recent research has identified new classes of "machine un-auditable" bugs that arise from both transactional contexts and source code. Existing detection methods require human understanding of underlying transaction logic and manual reasoning across different sources of context (i.e. modalities), such as code, dynamic transaction executions, and natural language specifying the expected transaction behavior. To automate the detection of ``machine un-auditable'' bugs, we present SmartInv, an accurate and fast smart contract invariant inference framework. Our key insight is that the expected behavior of smart contracts, as specified by invariants, relies on understanding and reasoning across multimodal information, such as source code and natural language. We propose a new prompting strategy to foundation models, Tier of Thought (ToT), to reason across multiple modalities of smart contracts and ultimately to generate invariants. By checking the violation of these generated invariants, SmartInv can identify potential vulnerabilities. We evaluate SmartInv on real-world contracts and re-discover bugs that resulted in multi-million dollar losses over the past 2.5 years (from January 1, 2021 to May 31, 2023). Our extensive evaluation shows that SmartInv generates (3.5X) more bug-critical invariants and detects (4$\times$) more critical bugs compared to the state-of-the-art tools in significantly (150X) less time. \sys uncovers 119 zero-day vulnerabilities from the 89,621 real-world contracts. Among them, five are critical zero-day bugs confirmed by developers as ``high severity.''

Cryptographic agility is gaining attention due to its crucial role in maintaining cryptographic security in a rapidly evolving technological landscape. However, despite its increasing importance, the term cryptographic agility remains vaguely defined and there is no clear consensus on its exact meaning. This lack of clarity poses a challenge since the need for agility becomes more urgent as new cryptographic vulnerabilities and advanced computing threats emerge, emphasizing the need for a systematic approach to clarify and refine the notion on cryptographic agility. In this paper, we systematize the concept of cryptographic agility by providing three research contributions. First, we review current definitions across academic and gray literature, identifying six distinct categories to differentiate every aspect within the definitions. Second, we synthesize these insights to establish a comprehensive, canonical definition of cryptographic agility. Third, we explore the relationship between cryptographic agility and the related concepts cryptographic versatility and interoperability. In our discussion, we examine the relevance of cryptographic agility, highlight its trade-offs with complexity, assess its individual applicability, and illustrate its various contexts by offering an additional application-specific definition. Our work provides a new perspective on cryptographic agility and related concepts, based on systematical research to clarify and enhance its future use., Comment: 18 pages, 1 figure

The evolution of wireless communication systems will be fundamentally impacted by an open radio access network (O-RAN), a new concept defining an intelligent architecture with enhanced flexibility, openness, and the ability to slice services more efficiently. For all its promises, and like any technological advancement, O-RAN is not without risks that need to be carefully assessed and properly addressed to accelerate its wide adoption in future mobile networks. In this paper, we present an in-depth security analysis of the O-RAN architecture, discussing the potential threats that may arise in the different O-RAN architecture layers and their impact on the Confidentiality, Integrity, and Availability (CIA) triad. We also promote the potential of zero trust, Moving Target Defense (MTD), blockchain, and large language models(LLM) technologies in fortifying O-RAN's security posture. Furthermore, we numerically demonstrate the effectiveness of MTD in empowering robust deep reinforcement learning methods for dynamic network slice admission control in the O-RAN architecture. Moreover, we examine the effect of explainable AI (XAI) based on LLMs in securing the system., Comment: 10 pages

Model Inversion (MI) attacks pose a significant threat to the privacy of Deep Neural Networks by recovering training data distribution from well-trained models. While existing defenses often rely on regularization techniques to reduce information leakage, they remain vulnerable to recent attacks. In this paper, we propose the Trapdoor-based Model Inversion Defense (Trap-MID) to mislead MI attacks. A trapdoor is integrated into the model to predict a specific label when the input is injected with the corresponding trigger. Consequently, this trapdoor information serves as the "shortcut" for MI attacks, leading them to extract trapdoor triggers rather than private data. We provide theoretical insights into the impacts of trapdoor's effectiveness and naturalness on deceiving MI attacks. In addition, empirical experiments demonstrate the state-of-the-art defense performance of Trap-MID against various MI attacks without the requirements for extra data or large computational overhead. Our source code is publicly available at https://github.com/ntuaislab/Trap-MID., Comment: Accepted by Neural Information Processing Systems (NeurIPS) 2024

An effective method for suppressing intentional forks in a blockchain is the last-generated rule, which selects the most recent chain as the main chain in the event of a chain tie. This rule helps invalidate blocks that are withheld by adversaries for a certain period. However, existing last-generated rules face an issue in that their applications to the system are not fully localized. In conservative cryptocurrency systems such as Bitcoin, it is desirable for methods to be applied in a fully local manner. In this paper, we propose a locally applicable last-generated rule. Our method is straightforward and is based on a relative time reference. By conservatively setting the upper bound for the clock skews $\Delta_{O_i}$ to 200 s, our proposed method reduces the proportion $\gamma$ of honest miners following the attacker during chain ties by more than 40% compared to existing local methods.

The vulnerability of Vision Large Language Models (VLLMs) to jailbreak attacks appears as no surprise. However, recent defense mechanisms against these attacks have reached near-saturation performance on benchmarks, often with minimal effort. This simultaneous high performance in both attack and defense presents a perplexing paradox. Resolving it is critical for advancing the development of trustworthy models. To address this research gap, we first investigate why VLLMs are prone to these attacks. We then make a key observation: existing defense mechanisms suffer from an \textbf{over-prudence} problem, resulting in unexpected abstention even in the presence of benign inputs. Additionally, we find that the two representative evaluation methods for jailbreak often exhibit chance agreement. This limitation makes it potentially misleading when evaluating attack strategies or defense mechanisms. Beyond these empirical observations, our another contribution in this work is to repurpose the guardrails of LLMs on the shelf, as an effective alternative detector prior to VLLM response. We believe these findings offer useful insights to rethink the foundational development of VLLM safety with respect to benchmark datasets, evaluation methods, and defense strategies.

The construction of attack technique knowledge graphs aims to transform various types of attack knowledge into structured representations for more effective attack procedure modeling. Existing methods typically rely on textual data, such as Cyber Threat Intelligence (CTI) reports, which are often coarse-grained and unstructured, resulting in incomplete and inaccurate knowledge graphs. To address these issues, we expand attack knowledge sources by incorporating audit logs and static code analysis alongside CTI reports, providing finer-grained data for constructing attack technique knowledge graphs. We propose MultiKG, a fully automated framework that integrates multiple threat knowledge sources. MultiKG processes data from CTI reports, dynamic logs, and static code separately, then merges them into a unified attack knowledge graph. Through system design and the utilization of the Large Language Model (LLM), MultiKG automates the analysis, construction, and merging of attack graphs across these sources, producing a fine-grained, multi-source attack knowledge graph. We implemented MultiKG and evaluated it using 1,015 real attack techniques and 9,006 attack intelligence entries from CTI reports. Results show that MultiKG effectively extracts attack knowledge graphs from diverse sources and aggregates them into accurate, comprehensive representations. Through case studies, we demonstrate that our approach directly benefits security tasks such as attack reconstruction and detection., Comment: 21 pages, 15 figures, 8 tables

As data-driven and AI-based decision making gains widespread adoption in most disciplines, it is crucial that both data privacy and decision fairness are appropriately addressed. While differential privacy (DP) provides a robust framework for guaranteeing privacy and several widely accepted methods have been proposed for improving fairness, the vast majority of existing literature treats the two concerns independently. For methods that do consider privacy and fairness simultaneously, they often only apply to a specific machine learning task, limiting their generalizability. In response, we introduce SAFES, a Sequential PrivAcy and Fairness Enhancing data Synthesis procedure that sequentially combines DP data synthesis with a fairness-aware data transformation. SAFES allows full control over the privacy-fairness-utility trade-off via tunable privacy and fairness parameters. We illustrate SAFES by combining AIM, a graphical model-based DP data synthesizer, with a popular fairness-aware data pre-processing transformation. Empirical evaluations on the Adult and COMPAS datasets demonstrate that for reasonable privacy loss, SAFES-generated synthetic data achieve significantly improved fairness metrics with relatively low utility loss.

AI-synthesized speech, also known as deepfake speech, has recently raised significant concerns due to the rapid advancement of speech synthesis and speech conversion techniques. Previous works often rely on distinguishing synthesizer artifacts to identify deepfake speech. However, excessive reliance on these specific synthesizer artifacts may result in unsatisfactory performance when addressing speech signals created by unseen synthesizers. In this paper, we propose a robust deepfake speech detection method that employs feature decomposition to learn synthesizer-independent content features as complementary for detection. Specifically, we propose a dual-stream feature decomposition learning strategy that decomposes the learned speech representation using a synthesizer stream and a content stream. The synthesizer stream specializes in learning synthesizer features through supervised training with synthesizer labels. Meanwhile, the content stream focuses on learning synthesizer-independent content features, enabled by a pseudo-labeling-based supervised learning method. This method randomly transforms speech to generate speed and compression labels for training. Additionally, we employ an adversarial learning technique to reduce the synthesizer-related components in the content stream. The final classification is determined by concatenating the synthesizer and content features. To enhance the model's robustness to different synthesizer characteristics, we further propose a synthesizer feature augmentation strategy that randomly blends the characteristic styles within real and fake audio features and randomly shuffles the synthesizer features with the content features. This strategy effectively enhances the feature diversity and simulates more feature combinations.

We introduce a set of useful expressions of Differential Privacy (DP) notions in terms of the Laplace transform of the privacy loss distribution. Its bare form expression appears in several related works on analyzing DP, either as an integral or an expectation. We show that recognizing the expression as a Laplace transform unlocks a new way to reason about DP properties by exploiting the duality between time and frequency domains. Leveraging our interpretation, we connect the $(q, \rho(q))$-R\'enyi DP curve and the $(\epsilon, \delta(\epsilon))$-DP curve as being the Laplace and inverse-Laplace transforms of one another. This connection shows that the R\'enyi divergence is well-defined for complex orders $q = \gamma + i \omega$. Using our Laplace transform-based analysis, we also prove an adaptive composition theorem for $(\epsilon, \delta)$-DP guarantees that is exactly tight (i.e., matches even in constants) for all values of $\epsilon$. Additionally, we resolve an issue regarding symmetry of $f$-DP on subsampling that prevented equivalence across all functional DP notions.

We explore the trade-off between privacy and statistical utility in private two-sample testing under local differential privacy (LDP) for both multinomial and continuous data. We begin by addressing the multinomial case, where we introduce private permutation tests using practical privacy mechanisms such as Laplace, discrete Laplace, and Google's RAPPOR. We then extend our multinomial approach to continuous data via binning and study its uniform separation rates under LDP over H\"older and Besov smoothness classes. The proposed tests for both discrete and continuous cases rigorously control the type I error for any finite sample size, strictly adhere to LDP constraints, and achieve minimax separation rates under LDP. The attained minimax rates reveal inherent privacy-utility trade-offs that are unavoidable in private testing. To address scenarios with unknown smoothness parameters in density testing, we propose an adaptive test based on a Bonferroni-type approach that ensures robust performance without prior knowledge of the smoothness parameters. We validate our theoretical findings with extensive numerical experiments and demonstrate the practical relevance and effectiveness of our proposed methods., Comment: 59 pages, 5 figures

Machine learning (ML) based indoor localization solutions are critical for many emerging applications, yet their efficacy is often compromised by hardware/software variations across mobile devices (i.e., device heterogeneity) and the threat of ML data poisoning attacks. Conventional methods aimed at countering these challenges show limited resilience to the uncertainties created by these phenomena. In response, in this paper, we introduce SAFELOC, a novel framework that not only minimizes localization errors under these challenging conditions but also ensures model compactness for efficient mobile device deployment. Our framework targets a distributed and co-operative learning environment that uses federated learning (FL) to preserve user data privacy and assumes heterogeneous mobile devices carried by users (just like in most real-world scenarios). Within this heterogeneous FL context, SAFELOC introduces a novel fused neural network architecture that performs data poisoning detection and localization, with a low model footprint. Additionally, a dynamic saliency map-based aggregation strategy is designed to adapt based on the severity of the detected data poisoning scenario. Experimental evaluations demonstrate that SAFELOC achieves improvements of up to 5.9x in mean localization error, 7.8x in worst-case localization error, and a 2.1x reduction in model inference latency compared to state-of-the-art indoor localization frameworks, across diverse building floorplans, mobile devices, and ML data poisoning attack scenarios.

The remarkable advances in deep learning have led to the emergence of many off-the-shelf classifiers, e.g., large pre-trained models. However, since they are typically trained on clean data, they remain vulnerable to adversarial attacks. Despite this vulnerability, their superior performance and transferability make off-the-shelf classifiers still valuable in practice, demanding further work to provide adversarial robustness for them in a post-hoc manner. A recently proposed method, denoised smoothing, leverages a denoiser model in front of the classifier to obtain provable robustness without additional training. However, the denoiser often creates hallucination, i.e., images that have lost the semantics of their originally assigned class, leading to a drop in robustness. Furthermore, its noise-and-denoise procedure introduces a significant distribution shift from the original distribution, causing the denoised smoothing framework to achieve sub-optimal robustness. In this paper, we introduce Fine-Tuning with Confidence-Aware Denoised Image Selection (FT-CADIS), a novel fine-tuning scheme to enhance the certified robustness of off-the-shelf classifiers. FT-CADIS is inspired by the observation that the confidence of off-the-shelf classifiers can effectively identify hallucinated images during denoised smoothing. Based on this, we develop a confidence-aware training objective to handle such hallucinated images and improve the stability of fine-tuning from denoised images. In this way, the classifier can be fine-tuned using only images that are beneficial for adversarial robustness. We also find that such a fine-tuning can be done by updating a small fraction of parameters of the classifier. Extensive experiments demonstrate that FT-CADIS has established the state-of-the-art certified robustness among denoised smoothing methods across all $\ell_2$-adversary radius in various benchmarks., Comment: 26 pages; TMLR 2024; Code is available at https://github.com/suhyeok24/FT-CADIS

We introduce LLMStinger, a novel approach that leverages Large Language Models (LLMs) to automatically generate adversarial suffixes for jailbreak attacks. Unlike traditional methods, which require complex prompt engineering or white-box access, LLMStinger uses a reinforcement learning (RL) loop to fine-tune an attacker LLM, generating new suffixes based on existing attacks for harmful questions from the HarmBench benchmark. Our method significantly outperforms existing red-teaming approaches (we compared against 15 of the latest methods), achieving a +57.2% improvement in Attack Success Rate (ASR) on LLaMA2-7B-chat and a +50.3% ASR increase on Claude 2, both models known for their extensive safety measures. Additionally, we achieved a 94.97% ASR on GPT-3.5 and 99.4% on Gemma-2B-it, demonstrating the robustness and adaptability of LLMStinger across open and closed-source models., Comment: Accepted at AAAI 2025

Fine-tuning large pre-trained foundation models (FMs) on distributed edge devices presents considerable computational and privacy challenges. Federated fine-tuning (FedFT) mitigates some privacy issues by facilitating collaborative model training without the need to share raw data. To lessen the computational burden on resource-limited devices, combining low-rank adaptation (LoRA) with federated learning enables parameter-efficient fine-tuning. Additionally, the split FedFT architecture partitions an FM between edge devices and a central server, reducing the necessity for complete model deployment on individual devices. However, the risk of privacy eavesdropping attacks in FedFT remains a concern, particularly in sensitive areas such as healthcare and finance. In this paper, we propose a split FedFT framework with differential privacy (DP) over wireless networks, where the inherent wireless channel noise in the uplink transmission is utilized to achieve DP guarantees without adding an extra artificial noise. We shall investigate the impact of the wireless noise on convergence performance of the proposed framework. We will also show that by updating only one of the low-rank matrices in the split FedFT with DP, the proposed method can mitigate the noise amplification effect. Simulation results will demonstrate that the proposed framework achieves higher accuracy under strict privacy budgets compared to baseline methods., Comment: 6 pages, 3 figures, submitted to IEEE ICC 2025

Certification Authority Authentication (CAA) is a safeguard against illegitimate certificate issuance. We show how shortcomings in CAA concepts and operational aspects undermine its effectiveness in preventing certificate misissuance. Our discussion reveals pitfalls and highlights best practices when designing security protocols based on DNS., Comment: 9 pages, 4 figures

Code Language Models (CLMs) have achieved tremendous progress in source code understanding and generation, leading to a significant increase in research interests focused on applying CLMs to real-world software engineering tasks in recent years. However, in realistic scenarios, CLMs are exposed to potential malicious adversaries, bringing risks to the confidentiality, integrity, and availability of CLM systems. Despite these risks, a comprehensive analysis of the security vulnerabilities of CLMs in the extremely adversarial environment has been lacking. To close this research gap, we categorize existing attack techniques into three types based on the CIA triad: poisoning attacks (integrity \& availability infringement), evasion attacks (integrity infringement), and privacy attacks (confidentiality infringement). We have collected so far the most comprehensive (79) papers related to adversarial machine learning for CLM from the research fields of artificial intelligence, computer security, and software engineering. Our analysis covers each type of risk, examining threat model categorization, attack techniques, and countermeasures, while also introducing novel perspectives on eXplainable AI (XAI) and exploring the interconnections between different risks. Finally, we identify current challenges and future research opportunities. This study aims to provide a comprehensive roadmap for both researchers and practitioners and pave the way towards more reliable CLMs for practical applications., Comment: Under a reviewing process since Sep. 3, 2024

Recent advances in voice synthesis, coupled with the ease with which speech can be harvested for millions of people, introduce new threats to applications that are enabled by devices such as voice assistants (e.g., Amazon Alexa, Google Home etc.). We explore if unrelated and limited amount of speech from a target can be used to synthesize commands for a voice assistant like Amazon Alexa. More specifically, we investigate attacks on voice assistants with synthetic commands when they match command sources to authorized users, and applications (e.g., Alexa Skills) process commands only when their source is an authorized user with a chosen confidence level. We demonstrate that even simple concatenative speech synthesis can be used by an attacker to command voice assistants to perform sensitive operations. We also show that such attacks, when launched by exploiting compromised devices in the vicinity of voice assistants, can have relatively small host and network footprint. Our results demonstrate the need for better defenses against synthetic malicious commands that could target voice assistants.

As businesses increasingly adopt cloud technologies, they also need to be aware of new security challenges, such as server-side script attacks, to ensure the integrity of their systems and data. These scripts can steal data, compromise credentials, and disrupt operations. Unlike executables with standardized formats (e.g., ELF, PE), scripts are plaintext files with diverse syntax, making them harder to detect using traditional methods. As a result, more sophisticated approaches are needed to protect cloud infrastructures from these evolving threats. In this paper, we propose novel feature extraction and deep learning (DL)-based approaches for static script malware detection, targeting server-side threats. We extract features from plain-text code using two techniques: syntactic code highlighting (SCH) and abstract syntax tree (AST) construction. SCH leverages complex regexes to parse syntactic elements of code, such as keywords, variable names, etc. ASTs generate a hierarchical representation of a program's syntactic structure. We then propose a sequential and a graph-based model that exploits these feature representations to detect script malware. We evaluate our approach on more than 400K server-side scripts in Bash, Python and Perl. We use a balanced dataset of 90K scripts for training, validation, and testing, with the remaining from 400K reserved for further analysis. Experiments show that our method achieves a true positive rate (TPR) up to 81% higher than leading signature-based antivirus solutions, while maintaining a low false positive rate (FPR) of 0.17%. Moreover, our approach outperforms various neural network-based detectors, demonstrating its effectiveness in learning code maliciousness for accurate detection of script malware.