Graph alignment, the task of identifying corresponding nodes across different graphs, is crucial for applications ranging from social network analysis to bioinformatics. Although most existing methods leverage graph neural networks (GNNs) to learn node embeddings for attributed graphs and match them based on node similarity, they often rely on objectives designed for node classification or link prediction. These approaches preserve node proximity within individual graphs but fail to capture cross-graph correspondence knowledge, leading to suboptimal alignment performance.

Non-Fungible Tokens (NFTs) enable the decentralized representation and exchange of real-world assets, supporting features like fractional ownership and programmable logic that underpin emerging digital finance ecosystems. However, the openness of NFT markets makes them susceptible to manipulation tactics like wash trading, where coordinated trades distort prices. This undermines valuation accuracy and erodes trust in decentralized finance. To counter these challenges, we propose NFTGuard, a unified framework for detecting wash trades and producing manipulation-resilient NFT price predictions. First, NFTGuard filters out manipulated transactions using a rule-based detector that identifies self-dealing and cyclic trading patterns prevalent in decentralized marketplaces. Second, NFTGuard prepares for price prediction by constructing a multi-modal representation that integrates temporal trading dynamics, transactional metadata, and asset-specific semantic signals. Third, NFTGuard performs prediction using a Multi-Layer Perceptron (MLP) mixing backbone that fuses these heterogeneous cues into manipulation-resilient forecasts. Experiments on real-world NFT datasets from platforms like Rarible and Opensea show that NFTGuard achieves a 90.9% F1-score in detecting wash trades and improves price prediction accuracy by over 10% compared to the baselines.

Establishing efficient and robust covert channels is crucial for secure communication within insecure network environments. With its inherent benefits of decentralization and anonymization, blockchain has gained considerable attention in developing covert channels. To guarantee a highly secure covert channel, channel negotiation should be contactlessbeforethe communication, carrier transaction features must be indistinguishable from normal transactionsduringthe communication, and communication identities must be untraceableafterthe communication. Such a full-lifecycle covert channel is indispensable to defend against a versatile adversary who intercepts two communicating parties comprehensively (e.g., on-chain and off-chain). Unfortunately, it has not been thoroughly investigated in the literature. We make the first effort to achieve a full-lifecycle covert channel, a novel blockchain-based covert channel namedABC-Channel. We tackle a series of challenges, such as off-chain contact dependency, increased masquerading difficulties as growing transaction volume, and time-evolving, communicable yet untraceable identities, to achieve contactless channel negotiation, indistinguishable transaction features, and untraceable communication identities, respectively. We develop a working prototype to validateABC-Channeland conduct extensive tests on the Bitcoin testnet. The experimental results demonstrate thatABC-Channelachieves substantially secure covert capabilities. In comparison to existing methods, it also exhibits state-of-the-art transmission efficiency.

Anonymous credential systems (ACs) enable users to selectively disclose attributes in an anonymous and unlinkable manner, but their misuse is hard to address without traceability. Existing traceable ACs introduce regulators to handle this issue, but they impose substantial communication and computational overhead on regulators, especially in complex scenarios involving large numbers of authentication records and multiple independent regulators. In this work, we propose Public-Key Encryption with Equality Test and Variable Public Generator (PKEET-VPG) and its verifiable version (VPV) for adapting to traceable ACs to support secure outsourceable record retrieval from regulators to service providers. The core technical idea is a session-user-level tracing key whose validity is bound to a single authentication session, thereby preventing the abuse of tracing keys.We formally prove that both schemes achieve OW-CCA2 security, and show that, when integrated into ACs, they preserve user anonymity, support non-frameability, and retain cross-session unlinkability. Theoretical analysis and comparative experiments demonstrate that our schemes can reduce the overhead of regulators at a moderate cost in the authentication phase. Furthermore, our schemes support session-level ciphertext deduplication, which may be of independent interest in some scenarios, such as anonymous voting or one-action-per-session authentication.

In an open and complex environment of drone service collaboration scenarios, efficient identity authentication is essential for secure service interactions. Traditional schemes suffer from high computational costs, low communication efficiency, and vulnerability to single-point failures. This paper presents a new lightweight authentication scheme (BDP-Auth) based on blockchain decentralized identity (DID) and physically unclonable functions (PUFs). PUF enables low-power, low-cost identity key generation without relying on complex computation or large storage, ensuring tamper-proof identifiers while reducing hardware and energy overhead. Using blockchain decentralization, the DID mechanism securely registers and stores drone identities, offering transparency and integrity. Security analysis shows that BDP-Auth achieves forward and backward security, meaning that the compromise of the current session key does not reveal past or subsequent session communications, and also resists physical capture and impersonation attacks. Experiments demonstrate that, compared to PRLAP-IoD, BDP-Auth reduces the computation overhead by 38.3% and 45.0% on the STM32F4 and NodeMCU platforms, respectively, decreases communication overhead by 36.9%, and reduces storage overhead by 27.5%, significantly improving overall efficiency.

Malware widely adopts network traffic encryption techniques to conceal malicious activities. Recent research has demonstrated the effectiveness of machine learning (ML)-, deep learning (DL)-, and pre-training-based malware traffic detection methods. However, a vast majority of these methods rely on the collected complete traffic during the malware attack. While certain methods can operate on partial traffic, their detection accuracy often significantly decreases when the available data is restricted to the extreme early stage, where information is most sparse. In this paper, we proposeDawnGuard, an effective early-stage encrypted malware traffic detection framework through multi-flow temporal graph learning. Specifically, based on the temporal packet density distribution analysis,DawnGuardinnovatively proposes a self-adjusting data augmentation strategy for early-stage malware traffic, which can force the model to focus on the early-stage interaction phase with more distinguishable properties. Meanwhile, considering that temporal-topological correlations among multiple flows can provide more distinguishable properties in a malware attack, we further develop a temporal graph learning framework to extract features, which can formMulti-Flow Graph Features (MGF). By utilizingMGF,Dawn-Guardimplements a Vision Transformer-based detection mechanism, enabling accurate and precise encrypted malware traffic detection with early-stage traffic by capturing both local and global contextual relationships. Extensive experiments with two real-world datasets demonstrate thatDawnGuardoutperforms the state-of-the-art (SOTA) methods in three typical scenarios: varying early-stage time windows, imbalanced data, and unseen malware detection. Particularly,DawnGuardachieves an average F1 of 95.11%, 8.7% higher than the SOTA method, by only utilizing the first 20% loading ratio of complete traffic.

With the rapid development of Large Language Models (LLMs), an increasing number of Large Visual-Language Models (LVLMs) have achieved unprecedented performance in response generation. Recent work shows that LVLMs are vulnerable to adversarial attacks. However, many existing methods tend to overfit to the source model by overemphasizing specific features, which compromises their transferability. Other approaches suffer from reduced attack effectiveness due to insufficient differentiation between features. In this paper, we propose a novel transfer-based black-box untargeted attack—Shared Adversarial Feature (SAF) dynamic attack. By exploring the feature extraction patterns of LVLMs, we identify the features shared among various models that are most susceptible to adversarial attacks and disrupt them. Moreover, due to the powerful attention mechanisms of LVLMs, they are still able to extract similar semantics from perturbed images, even when primary features are disrupted. We design a dynamic update strategy to address this challenge. Finally, from the perspective of SAF, we conduct an in-depth analysis of vulnerabilities in the vision encoder and projector within LVLMs and find that attacking the projector exhibits stronger transferability across heterogeneous model architectures. Extensive experiments show that our method exhibits superior attack performance compared to existing methods across different models, datasets, and tasks. The code will be publicly available after publication.

Vertical Split Federated Learning (VSFL) allows participants to collaboratively train a better model with different features vertically partitioned in the same sample space, where the model is divided into bottom model and top model by the cut layer, trained by passive and active participants respectively. However, in the process, the labels owned by the active participant will still be inferred or stolen by curious or malicious passive participants. In this paper, we propose Casper, a causality-inspired defense mechanism with a confounder against label inference attacks in VSFL. Casper first analyzes the feasibility of optimizing the training process in VSFL at the intervention level from a causal perspective. It then introduces a confounder consisting of cut layer output reconstruction and label obfuscation to disrupt the direct causality between cut layer outputs and labels. Additionally, we integrate selective discrepancy training to further ensure model utility by strategically balancing training between active and passive participants. Extensive experiments conducted on four datasets across different tasks demonstrate that Casper effectively preserves label privacy while maintaining model performance, significantly outperforming current advanced defending methods in VSFL.

The widespread deployment of Internet of Things (IoT) devices has driven their segmentation into distinct trust domains for the purpose of governance, creating a critical need for secure cross-domain authentication (CDA). CDA must preserve both anonymity and traceability of device identities to enable trustworthy data exchange. However, existing approaches, while exploring this trade-off, remain vulnerable to single points of failure and Sybil attacks—threats that are especially severe for unattended and resource-constrained devices. In this paper, we propose a Self-Sovereign and Supervised Cross-domain authentication scheme (SCross) to tackle these issues. The main building block we designed is a pseudonym management scheme (PMS) that allows devices to generate and use pseudonyms without relying on a trusted party. Although devices has full control of their identities, PMS still ensures traceability, Sybil resistance, and revocability. We define the formal security models of PMS, instantiate it under two different approaches, namely group signature (SCross-GS) and zero-knowledge succinct non-interactive arguments of knowledge (zkSNARKs, SCross-ZK), and present security proofs for our proposal. We implemented and evaluated SCross. The result shows that our scheme achieves an effective trade-off between security and efficiency.

Security vulnerabilities in smart contracts can have severe economic consequences. Existing smart contract vulnerability detection methods rely primarily on rigid rules defined by experts and have difficulty in detecting unknown vulnerabilities. This article proposes a new Anomalous Smart Contract Detector, named ASCD, to effectively detect known and unknown vulnerabilities in smart contracts. This is achieved by interpreting unknown vulnerabilities as code anomalies and detecting them with an anomaly detection technique named DeepSVDD. This is also attributed to a new design of feature extraction, in which we compile smart contract source codes into opcodes, extract semantic features from opcode sequences, and control flow features from control flow graphs. By joining LSTM and GIN, the semantic and control flow features are fused to offer a comprehensive representation of smart contracts suitable for anomaly detection. Extensive experiments were conducted to verify the ASCD model, and more than 30,000 smart contracts were tested. The new model offers a significantly better F1-score than existing methods in detecting known vulnerabilities and achieves a high accuracy of 77% in detecting unknown vulnerabilities.

Federated learning significantly enhances immersive user experiences in the human-centric metaverse by training machine learning models while keeping distributed data localized. As the right to be forgotten is legislated globally, users in the human-centric metaverse should have the option to have their data forgotten. Federated Unlearning (FU) is an emerging solution that enables the server to remove client contributions from trained global models by calibrating the historical model parameters of participating clients. However, retaining the entire model parameters incurs significant storage overhead, and the involvement of the remaining clients exposes the global model to the risk of backdoor attacks. This paper presents a new parameter-efficient and backdoor-resistant FU framework, called FedHydra. FedHydra segments the model into a base and a classifier layer. By storing the classifier parameters critical to predicting output, FedHydra reduces the storage overhead of the server. FedHydra also calibrates the classifier parameters and compares cosine similarity between the old and new classifier parameters of the remaining clients, reducing the computational overhead of unlearning while effectively preventing the abnormal clients from participating in the calibration. Experiments show that, compared with the full parameter-based storage calibration method, FedHydra achieves comparable unlearning effects, re duces storage overhead on the server, and improves unlearning efficiency. In addition, FedHydra reduces the attack success rates in different backdoor attack settings while maintaining the predictive performance of the global model.

Blockchains increasingly enable decentralized and trustworthy service execution. Smart contracts, i.e., self-executing programs on the blockchain, are essential for automating services. Yet, their inherent vulnerabilities can severely compromise the integrity and reliability of services. Existing detection methods are largely based on expert-defined rules, limiting their effectiveness to known vulnerability types and struggling with anomalous contracts that exhibit variant behaviors. Data imbalance further hinders the performance of deep learning-based approaches. This article presents TLSCG, a transfer learning-based framework that enhances smart contract vulnerability detection, particularly for unknown vulnerabilities, by leveraging a LSTM-based Variational Autoencoder (VAE) to enrich the training set with diverse vulnerable samples. TLSCG enhances the realism of generated contracts by incorporating a Bigram loss to preserve local semantic coherence and a Generative Adversarial Network (GAN)-based discriminator to ensure global structural consistency. Using transfer learning, the generative model can adapt to new vulnerability types, enriching training data and improving detector generalization. We further present OpTrans, a semantic- and structure-aware Transformer optimized for opcode sequence modeling. By integrating type-guided embeddings with a sparse and structure-aware attention mechanism, OpTrans effectively captures instruction semantics and execution structure. Experiments on real-world datasets show that TLSCG achieves 84% Macro-F1 in detecting unknown vulnerabilities, a 19% improvement over the state-of-the-art method Escort. Evaluations of generated samples in realism, anomalousness, and discrepancy confirm their value in improving detection.

Telecommunication fraud refers to deceptive activities in the field of communication services. This research focuses on a category of fraud identified as ''mutation telecommunication fraud". There is currently a lack of research on mutation telecommunication fraud detection, allowing this type of fraud to persist uncaught. We identify that detecting mutation fraud requires capturing multi-source patterns, including user communication graphs and temporal-spatial Voice of Call (VOC) features. Specifically, we introduce MutationGuard, which leverages Graph Neural Networks (GNN) to capture changes in user communication graphs. For VOC records, we map call start times onto a 3D cylindrical surface, thereby representing each VOC record in spatial coordinates and utilizing proposed LFFE and TCFE modules to capture local fraud behaviors and temporal behavior changes. The proposed neural modeling approach that facilitates multi-source information fusion constitutes a significant advancement in detecting mutation fraud. Experiment results reveal a significant improvement in the AUC score by 1.52% and the F1 score by 1.36% on the proposed telecommunication fraud dataset. Particularly, our method shows a significant improvement of 13.93% in accuracy on mutation fraud data. We also validate the effectiveness of our method on the publicly available Sichuan Telecommunication Fraud dataset.

Existing research indicates that original federated learning is not absolutely secure; attackers can infer the original training data based on reconstructed gradient information. Therefore, we will further investigate methods to protect data privacy and prevent adversaries from reconstructing sensitive training samples from shared gradients. To achieve this, we propose a defense strategy called SLGD, which enhances model robustness by combining sparse learning and gradient perturbation techniques. The core idea of this approach consists of two parts. First, before processing training data at the RSU, we preprocess the data using sparse techniques to reduce data transmission and compress data size. Second, the strategy extracts feature representations from the model and performs gradient filtering based on the l 2 norm of this layer. Selected gradient values are then perturbed using Von Mises–Fisher (VMF) distribution to obfuscate gradient information, thereby defending against gradient reconstruction attacks and ensuring model security. Finally, we validate the effectiveness and superiority of the proposed method across different datasets and attack scenarios.

In the Social Internet of Things (SIoT), edge computing integrates artificial intelligence to learn intricate relationships. The scale and complexity of SIoT cause a data explosion from diverse objects, hindering tailored services to users who own objects. Moreover, conventional edge computing in SIoT depends on centralized data collection, raising concerns about data privacy. To address the above two issues, federated recommender systems (FRSs) present a promising solution. FRSs can provide SIoT services to users and train a shared model while retaining sensitive data locally on objects. However, as FRSs are driven by data, they are inherently susceptible to algorithmic bias, raising substantial fairness concerns that have attracted considerable attention in SIoT. Recent fairness studies predominantly concentrate on a single sensitive attribute for users, thereby overlooking their autonomy. Therefore, we propose CustomFair, a personalized fairness framework that enables users in FRSs to select preferred sensitive attributes and acquire satisfied recommendation services in SIoT scenarios. First, we define customized fairness to ensure group fairness based on users’ sensitive attributes. The server segments users into subgroups in a privacy-preserving manner. Second, CustomFair employs the DynBalance method with a flexible regularization coefficient to improve recommendation performance and utilizes the AdaptEpoch strategy to achieve fairness. Extensive experiments indicate that CustomFair improves recommendation performance by 0.1–42.92 and enhances fairness by reducing disparities of 0.03–5.41 compared to two baselines across three datasets.

With the proliferation of data sharing in the Internet of Things (IoT), protecting privacy-sensitive information and preventing unauthorized access have become paramount concerns. Existing centralized access control methods faces single-point-of-failure risks and lacks scalability for dynamic IoT systems. This paper proposes a fine-grained access control framework based on cross-blockchain technology for transparent and flexible IoT data sharing. In our framework, the cross-blockchain module is facilitated to eliminate data isolation across domains, the Interplanetary File System (IPFS) is used to mitigate centralized storage risks and reduce blockchain storage overhead, CP-ABE and symmetric encryption are integrated to enforce attribute-based, fine-grained access control with strong security guarantees. Meanwhile, the blockchain records data and key references to enforce secure access to shared data content. We further conduct security analysis and experimental evaluations, demonstrating the effectiveness and efficiency of the proposed scheme.

With the rapid advancement of wearable devices and Internet of Things (IoT) technologies, sensor data generated by edge devices has surged. This data is crucial for advancing IoT applications, including health status monitoring, abnormal behavior detection, and environmental monitoring. However, traditional centralized learning requires uploading data to a central server, raising security and privacy concerns and hindering data application. Federated learning (FL) offers a solution by enabling collaborative model training on IoT devices without transferring data from the local device. In practice, edge devices generate data that is often highly heterogeneous, making it challenging for the global FL model to capture local data distributions accurately, leading to significant performance degradation. Additionally, imbalanced edge device resources and limited bandwidth can cause data transmission delays or interruptions, impacting application feasibility. To address these issues, we propose PFedKD, a novel personalized FL algorithm based on knowledge distillation, aimed at enhancing the model’s generalization ability and reducing communication overhead in heterogeneous IoT data environments. PFedKD constructs a public dataset using unlabeled pseudo data to extract knowledge from each client, training personalized models that fit local data distributions. This method controls dataset size while enhancing performance. During communication, only logits and class prototypes are transmitted, ensuring high communication efficiency. Sharpness aware minimization is introduced in local model training to optimize generalization. Additionally, we design a weight distribution mechanism based on client sample quality evaluation that optimizes knowledge aggregation and model personalization. Extensive experiments demonstrate that PFedKD significantly outperforms state-of-the-art baselines in both learning performance and communication efficiency.

Performing effective supervision in blockchain networks while preserving public anonymity has long been a challenging issue. Existing solutions often depend on third-party institutions or dual-chain architectures to monitor and recover user identities, but these approaches pose significant risks, such as identity disclosure and increased inter-chain communication costs. To address these concerns, this paper proposes a novel supervised scheme based on Threshold Secret-Sharing (TSS) and Zero-Knowledge Proof (ZKP) on anonymous accounts on the blockchain called TSS-ZKP, while integrating regulatory authority (RA) and multiple traceability centers (TC). Compared to traditional single RA, TSS-ZKP decentralizes the regulatory authority and storage across multiple TCs, significantly reducing the risk of user identity leakage by a single RA. By TSS-ZKP, user identities are recovered through the collaborative generation of sub-secrets by the RA and TCs, without storing actual identity information. By leveraging elliptic curves and using hash functions as secret labels, TSS-ZKP achieves lightweight operations, allowing TCs to efficiently locate sub-secrets without extensive traversal. The comprehensive analysis of security and privacy demonstrated that TSS-ZKP effectively safeguards user privacy while enabling feasible supervision. In addition, comparison experimental results show that the time consumption of the TSS-ZKP is about 30% of that of the comparison scheme. Meanwhile, the simulation results highlight the practicality of the scheme, showing that TSS-ZKP significantly reduces the delays in the identity recovery process while maintaining high usability. In general, TSS-ZKP provides a safer and more feasible solution for enabling the supervision of blockchain accounts.

Vertical federated learning (VFL) shows promise for enabling collaborative learning among Internet of Vehicle systems (IoVs) without requiring the sharing of private training data. However, existing work has exposed VFL’s vulnerability to privacy-stealing attacks, where an honest but curious server might reconstruct a client’s raw data from client-uploaded embeddings. In this work, we first elucidate the intrinsic mechanisms of privacy attacks from an information theory perspective, which provides a solid foundation for potential defensive strategies. Based on our findings, we introduce PriVFL, a defense mechanism based on information bottleneck theory. PriVFL is designed to safeguard the privacy of VFL-based IoVs by enabling shared embeddings to extract minimal information from input data, while preserving the information essential to target labels. Specifically, PriVFL restricts the information contained in embeddings by reducing the upper bound of mutual information between the raw samples and embeddings uploaded from local clients. Meanwhile, PriVFL ensures the effectiveness of the model by increasing the mutual information lower bound between embeddings and samples’ labels. Our evaluation includes 5 benchmark data sets and 4 different models. Experimental results demonstrate that PriVFL effectively mitigates privacy attacks while preserving the model’s effectiveness. These findings underscore that PriVFL can significantly enhance the privacy of VFL-based IoVs, thereby bolstering the development of practical IoV applications.

Deep neural networks (DNNs) find extensive applications, including object detection in various security domains. However, these DNN models are susceptible to backdoor attacks. While significant research has been conducted on backdoor attacks in classified models, limited attention has been given to object detection models. Previous studies have predominantly focused on backdoor attacks in digital environments, overlooking real-world implications. Notably, the efficacy of backdoor attacks in real-world scenarios can be significantly influenced by physical factors such as distance and illumination. In this article, we introduce a variable-size backdoor trigger designed to accommodate objects of different sizes, mitigating disruptions arising from varying distances between the viewing point and the targeted object. Additionally, we propose malicious adversarial training for backdoor training, enabling the backdoor object detector to learn trigger features amidst physical noise. Experimental results demonstrate that our robust backdoor attack (RBA) enhances the success rate of attacks in real-world settings.

Ensuring fairness in incentive mechanisms for federated learning (FL) is essential to attracting high-quality clients and building a sustainable FL ecosystem. Most existing fairness-aware incentive mechanisms distribute rewards to FL clients by quantifying their contributions to the performance of the global model. Essentially, these mechanisms pursuecontribution fairness, namely a constant contribution-reward ratio across FL clients, with an implicit assumption that clients would be satisfied with thecontribution fairness. However, research in social psychology has confirmed that this assumption may not hold in many real-world scenarios. According to equity theory proposed by Adams, an individual’s assessment and perception of receiving fair treatment significantly depend on the input-outcome ratio, where outcome simply refers to the rewards, while input is far more complex because it involves a bunch of subtle factors such as enthusiasm, experience and tolerance as well as the estimated contributions. Inspired by Adams’ equity theory, in this work, we expand the notion ofcontribution fairnesstoinput fairnessand propose a new fairness-aware incentive mechanism namedFairRewardthat distributes rewards under the joint consideration of self-reported inputs and computed contributions.FairRewardemploys a reputation mechanism to enhance the credibility of self-reported inputs and leverages blockchains to eliminate the need of a trusted FL server and monetarily incentivize/penalize clients. In addition,FairRewardadopts techniques including distributed differential privacy and locality-sensitive hashing to address privacy and non-IID issues in FL. Moreover, we conduct a comprehensive security and privacy analysis. Finally, we evaluateFairRewardthrough extensive experiments. The comprehensive experimental results demonstrate thatFairRewardis effective, scalable and attack-resistant, and provides theinput fairnessrequired.

Conventional federated learning (FL) coordinated by a central server focuses on training a global model and protecting the privacy of clients' training data by storing it locally. However, the statistical heterogeneity hinders the global model from adapting to the non-IID distributions among clients. Moreover, untrusted and unreliable central servers and malicious clients may compromise model integrity and availability, thus degrading the robustness of FL. To address these challenges, we present RobustPFL, a decentralized personalized federated learning (PFL) approach that combines$\alpha$-based Layer-position Normalized Similarity ($\alpha$-LNS) and local collaborative training to improve personalized performance while utilizing a blockchain-based committee mechanism to coordinate the aggregation process, thereby achieving high personalized accuracy and robustness. Extensive experiments show that our RobustPFL approach outperforms multiple algorithms, including Local training, FedAvg, FedReptile, Per-FedAvg, FedBN, and SPFL, on MNIST, CIFAR10, EMNIST, and N-BaIoT datasets in four non-IID settings. We also evaluate RobustPFL's effectiveness against attacks—poisoning attacks and free-riding attacks. Particularly, for three prevalent poisoning attacks (backdoor, label flipping, and model poisoning attacks), we compare non-defensive (FedAvg) and defensive (Krum, trimmed mean, Bulyan, FedBN, FLAME, and FangTrmean) methods with our proposed RobustPFL. The results show that our approach achieves significant defensive effects.

Federated Learning is a data privacy-protected distributed machine learning framework, but malicious clients can damage it. Byzantine-robust federated learning aims to learn an accurate global model despite the presence of malicious clients. Most current defenses assume that clients have identically and independently distributed (i.i.d.) data and thus cannot perform well in canonical non-i.i.d. scenarios. Several non-i.i.d. statistic-based defenses have been recently proposed to identify malicious clients through gradient statistics without any auxiliary techniques. They thus can only ensure robustness under certain attacks with specific characteristics. We propose Sophon, a comprehensive defense using auxiliary data to combat an arbitrary number of malicious clients in both i.i.d. and non-i.i.d. cases. Specifically, Sophon first normalizes received client gradients to reduce the dominance of malicious gradients. Then it introduces a dual trust mechanism to assign the aggregation weight for each gradient. The dual trust mechanism estimates the consistency-based and diversity-based trust scores of client gradients and integrates the two scores as the aggregation weight to effectively suppress the impact of malicious gradients. Extensive experimental results on three datasets from different domains, with diverse models and FL scenarios, show that Sophon is robust in maintaining the overall accuracy of the training model.

Considerable efforts have been devoted to addressing distributed backdoor attacks in federated learning (FL) systems. While significant progress has been made in enhancing the security of FL systems, our study reveals that there remains a false sense of security surrounding FL. We demonstrate that colluding malicious participants can effectively execute backdoor attacks during the FL training process, exhibiting high sparsity and stealthiness, which means they can evade common defense methods with only a few attack iterations. Our research highlights this vulnerability by proposing aCollusiveBackdoorAttack namedCoBA.CoBAis designed to enhance the sparsity and stealthiness of backdoor attacks by offering trigger tuning to facilitate learning of backdoor training data, controlling the bias of malicious local model updates, and applying the projected gradient descent technique. By conducting extensive empirical studies on 5 benchmark datasets, we make the following observations: 1)CoBAsuccessfully circumvents 15 state-of-the-art defense methods for robust FL; 2) Compared to existing backdoor attacks,CoBAconsistently achieves superior attack performance; and 3)CoBAcan achieve persistent poisoning effects through significantly sparse attack iterations. These findings raise substantial concerns regarding the integrity of FL and underscore the urgent need for heightened vigilance in defending against such attacks.

Vision-language pretraining (VLP) models have demonstrated outstanding performance in image-text understanding tasks but remain highly susceptible to transferable adversarial attacks. While ensemble-based guided attacks improve adversarial transferability by increasing the diversity of image-text pairs, they primarily rely on spatial-domain data augmentation, which can lead to model overfitting to image details and limit the generalization capability of attacks. To address this limitation, this study proposes a frequency-domain adjustment-based adversarial attack method that modifies specific frequency components of input images to reduce detail interference and enhance the stability of adversarial examples. Additionally, a fine-grained feature extraction technique is introduced to optimize image-text alignment, further improving the transferability of cross-modal attacks. Experimental results demonstrate that the proposed method achieves superior attack transferability and generalization performance across two major VLP architectures, fusion models and alignment models, as well as multiple tasks on the Flickr30 K and MSCOCO datasets.

Natural language processing (NLP) models are widely used in various scenarios, yet they are vulnerable to adversarial attacks. Existing works aim to mitigate this vulnerability, but each work targets a specific attack category or has computational overhead limitations, making them vulnerable to adaptive attacks. In this paper, we exhaustively investigate the adversarial attack algorithms in NLP and discover that existing attack algorithms mainly disrupt the importance distribution of words in a text. A well-trained model can distinguish subtle importance distribution differences between clean and adversarial texts. Based on this intuition, we propose TextDefense, a new adversarial example detection framework that utilizes the target model’s capability to defend against adversarial attacks, requiring no prior knowledge. Unlike previous approaches, TextDefense is attack-type agnostic and outperforms existing methods in experiments with different architectures, datasets, and attack methods. We also discover that the target model’s generalizability is a leading factor influencing the performance of TextDefense. Finally, we provide insights into the adversarial attacks in NLP and the principles of our defense method by analyzing the properties of the target model and the adversarial example.

Vertical Federated Learning (VFL) is a collaborative learning paradigm where participants share the same sample space while splitting the feature space. In VFL, local participants host their bottom models for feature extraction and collaboratively train a classifier by exchanging intermediate results with the server owning the labels. Both local training data and bottom models contain privacy-sensitive information and are considered the intellectual property of each participant, and thus should be protected by the design of VFL. Our study exposes the fundamental susceptibility of VFL systems to privacy leaks, which arise from the collaboration between the server and clients during both training and testing. Based on our findings, we proposePISTE, a model-agnostic framework of privacy stealing attacks against VFL. PISTE delivers three privacy inference attacks, i.e., model stealing, data reconstruction, and property inference attacks on five benchmark datasets and four different model architectures. We further discuss four potential countermeasures. Experimental results show that all of them cannot prevent all three privacy stealing attacks in PISTE. In summary, our study demonstrates the inherent yet rarely uncovered vulnerability of VFL on leaking data and model privacy.

With the rapid evolution of Web3.0, cryptocurrency has become a cornerstone of decentralized finance. While these digital assets enable efficient and borderless financial transactions, their pseudonymous nature has also attracted malicious activities such as money laundering, fraud, and other financial crimes. Effective detection of malicious accounts is crucial to maintaining the security and integrity of the Web 3.0 ecosystem. Existing malicious account detection methods rely on large amounts of labeled data and suffer from low generalization. Label-efficient and generalizable malicious account detection remains a challenging task. In this paper, we propose ShadowEyes, a framework for detecting malicious accounts by leveraging interaction feature learning with only a small labeled dataset. Specifically, We first propose a generalized account representation named TxGraph, which captures the universal interaction features of Ethereum and Bitcoin. Then we carefully design an account representation augmentation method tailored to simulate the evolution of malicious accounts to generate positive pairs. We conduct extensive experiments using public datasets to evaluate the performance of ShadowEyes. The results demonstrate that it outperforms state-of-the-art (SOTA) methods in four typical scenarios. Specifically, in the scenario of acrossplatform malicious account detection, ShadowEyes maintains an F1 score of around 90%, which is 10% higher than the SOTA method. In the zero-shot learning scenario, it can achieve an F1 score of 79.56% for detecting gambling accounts, surpassing the SOTA method by 10.44%.

Ethereum serves as the cornerstone for value transfer in Web 3.0, providing a decentralized and efficient trust mechanism for global connectivity. However, the anonymity of Ethereum undermines market regulatory capabilities, leading to frequent malicious behaviors such as Ponzi Scheme, Money Laundering, and Phishing. Therefore, in the face of the diverse and continuously emerging malicious behaviors, implementing fine-grained detection is crucial for maintaining the prosperous development of the blockchain ecosystem. In this paper, we propose FiMAD, a fine-grained and class-incremental malicious account detection framework based on dynamic graph learning. Specifically, we first propose a general graph structure calledDynamic Account Relation Graph (DARG), which dynamically models Ethereum accounts from a continuous-time perspective. Then, we design a cascade graph feature extraction method to capture deep temporal evolution patterns and neighbor interaction features in DARG. Next, we construct a pre-training universal encoder to transform account features into high-dimensional embeddings, followed by fine-tuning the model classifier with a few labeled samples, enabling accurate fine-grained detection and rapid updates for incremental classes. We conduct extensive experiments using real Ethereum data. The results demonstrate that FiMAD outperforms state-of-the-art (SOTA) methods in fine-grained detection across five typical scenarios: class-incremental, full data, new malicious accounts, imbalanced data, and binary classification. In the class-incremental scenario, FiMAD improves the Macro-F1 by up to 26.4% compared to SOTA methods.

Fingerprinting is a network reconnaissance technique utilized for gathering information about online computing systems, including operation systems and applications. Unfortunately, attackers typically leverage fingerprinting techniques to locate, enumerate, and subsequently target vulnerable systems, which is the first primary stage of a cyber attack. In this work, we explore the susceptibility of machine learning (ML)-based classifiers to misclassification, where a slight perturbation in the packet is included to spoof OS fingerprints. We propose SOFI (Spoof OS Fingerprints), an adversarial example generation algorithm under TCP/IP specification constraints, to create effective perturbations in a packet for deceiving an OS fingerprint. Specifically, SOFI has three major technical innovations: (1) it is the first to utilize adversarial examples to automatically perturb fingerprinting techniques; (2) it complies with constraints and integrity of network packets; (3) it achieves a high success rate in spoofing OS fingerprints. We validate the effectiveness of adversarial packets against active and passive OS fingerprints, verifying the transferability and robustness of SOFI. Comprehensive experimental results demonstrate that SOFI automatically identifies applicable and available OS fingerprint features, unlike existing tools relying on expert knowledge.

In the era where Web3.0 values data security and privacy, adopting groundbreaking methods to enhance privacy in recommender systems is crucial. Recommender systems need to balance privacy and accuracy, while also having the ability to overcome cold start problems. The Differential Trust Mechanism (DTM) introduced in this paper is such an approach. The DTM provides a unique use of Gaussian distributions in modeling trust relationships within data, offering a novel way to balance recommendation accuracy with user privacy. This mechanism innovatively applies differential privacy principles, using Gaussian noise addition to protect individual user data from inference attacks, while maintaining the integrity and utility of the overall dataset. Unlike traditional anonymization techniques that often compromise data utility or vulnerability to reverse engineering, DTM provides a robust solution by dynamically adjusting privacy levels based on the trustworthiness of data requests. By combining DTM with existing mainstream recommendation algorithms, the prediction accuracy of MAE and RMSE increases by at least 6.60% and 2.69%, respectively. This dual benefit positions DTM as a significant advancement in secure data processing, especially relevant for online businesses and platforms where personalized recommendations are crucial yet privacy concerns are paramount.

Vertical Federated Learning (VFL) is susceptible to various one-party hijacking attacks, such as Replay and Generation attacks, where a single malicious client can manipulate the model to produce attacker-specified results, thereby compromising its reliability in real-world deployments. In this paper, we first uncover the underlying mechanisms of these attacks and observe that successful attacks induce significant discrepancies in the embedding-label associations across different clients. We establish a theoretical framework demonstrating how these discrepancies can serve as reliable indicators for detecting hijacking attempts. Building upon this insight, we propose VFLMonitor, a robust defense mechanism that leverages these embedding-label discrepancies to detect and mitigate hijacking attacks. Specifically, VFLMonitor identifies suspicious queries by analyzing differences in label estimations from multiple clients and applies a majority voting rule to correct or filter out these malicious queries. Moreover, VFLMonitor introduces a novel regularization strategy during training to reduce intra-class variance in embeddings, thereby enhancing their discriminative power and improving defense effectiveness. Extensive experi21 ments were conducted on 5 real-world datasets against 2 different attack types under 3 attack scenarios. The results demonstrate that VFLMonitor can effectively identify and exclude potential hijacked requests in all types of one-party hijacking attacks, while maintaining a meager false positive rate for legitimate queries.

In cooperative intelligent transportation systems (CITS), federated learning enables vehicles to train a global model without sharing private data. However, the lack of an unlearning mechanism to remove the influence of vehicle-specified data from the global model potentially violates data protection regulations regarding the right to be forgotten. While the existing federated unlearning (FU) methods exhibit promising unlearning effects, their practicality in CITS is hindered due to the time-consuming retraining steps required by other vehicles and the non-negligible performance sacrifice on the un-forgotten data. Therefore, achieving effective unlearning without extensive retraining, while minimizing performance degradation on the un-forgotten data remains a challenge. In this work, we propose FedEditor, an efficient and effective FU framework in CITS that addresses the above challenge by reconfiguring the global model’s representation space to remove critical classification-related knowledge from the unlearned data. Firstly, FedEditor enables vehicles to perform the unlearning process locally on the global model, eliminating the participation of other vehicles and improving efficiency. Secondly, FedEditor captures and aligns the representations of the unlearned data with those of the nearest incorrect class centroid derived from non-training data, ensuring effective unlearning while preserving the un-forgotten data’s knowledge relatively intact for achieving competitive model performance. Finally, FedEditor refines the global model’s output distributions using the vehicles’ remaining data and incorporates a drift-mitigating regularization term, minimizing the negative impact of unlearning operations on model performance. Experimental results show that FedEditor reduces the unlearning rate by up to 99.64% without time-consuming retraining, while limiting the predictive performance loss of the resulting global model to less than 3.88% across five models and seven datasets.

Deep neural networks (DNNs) are vulnerable to adversarial examples crafted by well-designed perturbations. This could lead to disastrous results on critical applications such as self-driving cars, surveillance security, and medical diagnosis. At present, adversarial training is one of the most effective defenses against adversarial examples. However, in traditional adversarial training, it is still difficult to achieve a good trade-off between clean accuracy and robustness since DNNs still learn spurious features. The intrinsic reason is that traditional adversarial training makes it difficult to fully learn core features from adversarial examples when noise and examples cannot be disentangled. In this paper, we disentangle the adversarial examples into natural and perturbed patterns by bit-plane slicing. We assume the higher bit-planes represent natural patterns and the lower bit-planes represent perturbed patterns, respectively. We propose Feature-Focusing Adversarial Training (F$^{2}$AT), which differs from previous work in that it enforces the model to focus on the core features from natural patterns and reduce the impact of spurious features from perturbed patterns. The experimental results demonstrated that the clean accuracy and adversarial robustness with our F$^{2}$AT can be significantly improved.

Federated Recommendation (FedRec) systems are recognized as privacy-preserving solutions for collaboratively training recommender models without sharing users’ private data. However, recent studies have revealed that FedRec systems are vulnerable to interaction-level membership inference attacks. In such attacks, a semi-honest server can employ crafted methods to infer users’ interacted items. In this article, we identify that user preference information is predominantly stored in the user-uploaded parameters rather than in the local parameters after local training. Leveraging this insight, we expose a new interaction vulnerability and introduce the PubPara attack. Our experiments show that PubPara improves the inference performance by at least 40% over existing attacks, while requiring minimal inference time and remaining robust against current defense methods. To safeguard user privacy without compromising recommender performance, we propose MultiVerse, a novel countermeasure. MultiVerse utilizes untrained items outside the user’s local training data to obfuscate the server’s inference of interacted items. It includes a four-step strategy (training, optimization, refinement, and denoising) to achieve robust defense. Extensive experiments on three representative FedRec models (F-NCF, F-LightGCN, and FedRAP) across three real-world datasets validate that MultiVerse significantly degrades the attack’s inference performance to near the level of random guess while maintaining lossless recommender performance.

A fundamental goal of Web 3.0 is to establish a decentralized network and application ecosystem, thereby enabling users to retain control over their data while promoting value exchange. However, the recent TRON-Steem takeover incident poses a significant threat to this vision. In this paper, we present a thorough empirical analysis of the TRON-Steem takeover incident. By conducting a fine-grained reconstruction of the stake and election snapshots within the Steem blockchain, one of the most prominent social-oriented blockchains, we quantify the marked shifts in decentralization pre and post the takeover incident, highlighting the severe threat that blockchain network takeovers pose to the decentralization principle of Web 3.0. Moreover, by employing heuristic methods to identify anomalous voters and conducting clustering analyses on voter behaviors, we unveil the underlying mechanics of takeover strategies employed in the TRON-Steem incident and suggest potential mitigation strategies, which contribute to the enhanced resistance of Web 3.0 networks against similar threats in the future. We believe the insights gleaned from this research help illuminate the challenges imposed by blockchain network takeovers in the Web 3.0 era, suggest ways to foster the development of decentralized technologies and governance, as well as to enhance the protection of Web 3.0 user rights.

Federated recommendation (FR) is a decentralised approach to training personalised recommender systems, protecting users' privacy by avoiding data collection. Despite its privacy advantages, FR remains vulnerable to poisoning attacks. We focus on untargeted poisoning attacks against FR which degrade the overall performance of recommender services, leading to a detrimental impact on user experience and service quality. In this paper, we propose a general framework to formalise untargeted attacks and identify the vital role played by the interplay between items and user profiles in determining FR's performance. We present an untargeted attack FRecAttack2 which exploits this interplay. Specifically, we develop various methods for sampling user profiles, which approximate user distributions with and without collusion among malicious users. Then we leverage a new measurement to identify items that can disrupt the original interplay with user profiles, based on the change velocity of items' recommendation scores during optimisation. Extensive experiments demonstrate the superiority of our attack, outperforming existing methods by up to 27.56%, and its stealthiness in evading mainstream defences. To counteract untargeted attacks, we present a defence GuardCQ to detect malicious users by quantifying their contribution to boost the right interplay between items and user profiles. Empirical results show that GuardCQ effectively mitigates the attack's impact on FR and enhances the robustness of FR against poisoning attacks.

Security vulnerabilities in smart contracts can have serious economic consequences. Existing smart contract vulnerability detection methods rely primarily on strict rules defined by experts, making current research limited to detecting specific known vulnerabilities and difficult to deal with other types of anomalous contracts (i.e., the variants of contracts with potentially known vulnerabilities). The imbalance of a smart contract dataset also affects the effectiveness of deep learning-based methods. This paper proposes a new transfer learning-based, anomalous smart contract generation (TLSCG) method for abnormal contract detection, aimed at effectively detecting known vulnerabilities and other anomalous contracts. This method trains a smart contract operation code sequence generation model and improves the authenticity of generating smart contracts by adding semantic regularization terms to the loss function. Through transfer learning, the generative model can be readily extended to new types of vulnerabilities, obtaining known vulnerabilities and anomalous contract generation models, expanding training data, improving the generalization of detection models, and enabling detection models to detect anomalous contracts. By using a real smart contract dataset for validation, experiments show that the proposed method can effectively improve the generalization of the model in detecting known vulnerabilities. Compared to the latest rule-based vulnerability detection tools, the accuracy of anomalous contract detection is improved by 40% and the F1 score is improved by 24%.

Edge intelligence enables resource-demanding Deep Neural Network (DNN) inference without transferring original data, addressing concerns about data privacy in consumer Inter-net of Things (IoT) devices. For privacy-sensitive applications, deploying models in hardware-isolated trusted execution environments (TEEs) becomes essential. However, the limited secure memory in TEEs poses challenges for deploying DNN inference, and alternative techniques like model partitioning and offloading introduce performance degradation and security issues. In this paper, we present a novel approach for advanced model deployment in TrustZone that ensures comprehensive privacy preservation during model inference. We design a memory-efficient management method to support memory-demanding inference in TEEs. By adjusting the memory priority, we effectively mitigate memory leakage risks and memory overlap conflicts, resulting in 32 lines of code alterations in the trusted operating system. Additionally, we leverage two tiny libraries: S-Tinylib (2,538 LoCs), a tiny deep learning library, and Tinylibm (827 LoCs), a tiny math library, to support efficient inference in TEEs. We implemented a prototype on Raspberry Pi 3B+ and evaluated it using three well-known lightweight DNN models. The experimental results demonstrate that our design significantly improves inference speed by 3.13 times and reduces power consumption by over 66.5% compared to non-memory optimization method in TEEs.

Graph Prompt Learning (GPL) bridges significant disparities between pretraining and downstream applications to alleviate the knowledge transfer bottleneck in real-world graph learning. While GPL offers superior effectiveness in graph knowledge transfer and computational efficiency, the security risks posed by backdoor poisoning effects embedded in pretrained models remain largely unexplored. Our study provides a comprehensive analysis of GPL's vulnerability to backdoor attacks. We introduce CrossBA, the first cross-context backdoor attack against GPL, which manipulates only the pretraining phase without requiring knowledge of downstream applications. Our investigation reveals both theoretically and empirically that tuning trigger graphs, combined with prompt transformations, can seamlessly transfer the backdoor threat from pretrained encoders to downstream applications.Through extensive experiments involving 3 representative GPL methods across 5 distinct cross-context scenarios and 5 benchmark datasets of node and graph classification tasks, we demonstrate that CrossBA consistently achieves high attack success rates while preserving the functionality of downstream applications over clean input. We also explore potential countermeasures against CrossBA and conclude that current defenses are insufficient to mitigate CrossBA. Our study highlights the persistent backdoor threats to GPL systems, raising trustworthiness concerns in the practices of GPL techniques.

Deep classification models have been widely utilized in Vehicular Road Cooperation. However, previous work indicates that deep classification models are vulnerable to the privacy risks of Membership Inference Attacks (MIAs). Most existing work of MIAs is based on two different assumptions. One assumes adversary-own shadow datasets with aligned tasks and distributions as private datasets, while this assumption necessitates that the adversary knows the distributions of private datasets. The other assumes adversary-own shadow datasets with distinct tasks and distributions from private datasets, while this assumption requires that the adversary knows the classification boundaries between members and non-members of the private dataset. Hence, these two assumptions do not always hold in real-world scenarios. In this work, we systematically assess the impact of adversaryown shadow datasets with aligned tasks but distinct distributions from private datasets on MIAs. These realistic shadow datasets acknowledge adversaries limited insights into the data distribution of the private dataset and the decision boundary between members and non-members. We divide these practical shadow datasets hosted by adversaries into 4 types: Data Noise, Label Noise, Imbalanced Data, and Cross-domain Data. We conduct extensive experiments with 7 prevalent MIAs and 4 types of shadow datasets. Experimental results mainly reveal two-fold findings. First, MIAs still maintain effectiveness using the shadow dataset with the aligned task but distinct distributions from the private dataset. Second, different levels of data distribution disparities manifest varying MIAs’ performances under certain types of shadow datasets.

Graph neural networks, effectively harnessing the extensive interactive data from Internet of Things (IoT) devices, significantly enhance service quality in IoT systems. However, traditional centralized training leads to the leakage of private data during the data collection and model training phases in IoT scenarios. Federated learning (FL) has emerged as a promising approach, facilitating collaborative model training across diverse IoT devices without sharing sensitive data. The intricate types and relationships among IoT devices from various institutions highlight the issues of class imbalance and graph heterogeneity across different clients. These issues decrease the performance of FL models. In this work, we focus on a more realistic scenario where the IoT institutions have only limited amount and types of data. We propose a cross-institutional federated heterogeneous graph learning method called FedHGL. It aims to mitigate the negative effects of class imbalance while maintaining the private data locally on clients during collaborative training. We employ a heterogeneous graph neural network as the training model for clients. FedHGL generates cross-client minority class samples to enhance the model performance. Additionally, it incorporates a compensation mechanism to prevent forgetting global information. FedHGL designs an adaptive aggregation coefficient that assigns weights to IoT institutions according to the class imbalance of their data, thereby optimizing the aggregation process. Extensive experiments demonstrate the effectiveness of FedHGL for class imbalance and heterogeneous graph data.

With the rapid development of decentralized financial (DeFi), the total value locked (TVL) in DeFi continues to increase. A big number of adversaries exploit logic vulnerabilities to attack DeFi applications for profit, such as flash loan attacks and price manipulation attacks. However, the current vulnerability detection tools for smart contracts cannot be directly used to detect the logic vulnerabilities generated by the combination of different protocols. How to characterize and detect DeFi attacks that exploited logic vulnerabilities is a big challenge. In this work, we propose a deep-learning-based attack detection system on DeFi, called DeFiScanner, in which we design a novel neural network that includes a global model, a local model, and a fusion model to characterize DeFi attacks. First, the unstructured emitted events are automatically and efficiently normalized. Second, the transaction-related features of normalized emitted events are enriched with the global model and the semantic features of emitted events are extracted with the local model. Finally, the transaction-related features and the semantic features of emitted events are fused efficiently with the fusion model to detect DeFi attacks. We collect a dataset that consists of 50 910 real-world DeFi transactions on Ethereum (ETH). The extensive experimental results demonstrate the effectiveness of DeFiScanner. The true positive rate (TPR) and the area under the receiver operating characteristic (ROC) curve of the system reach 0.91 and 0.97, respectively.

While vulnerability databases are important sources of information for software security, it is known that information in these databases is inconsistent. How to rectify these incorrect data is a challenging issue. In this article, we employ developer logs and patches to automatically identify vulnerable source code versions that each CVE really affects. Our tool organizes all versions of a piece of software into a version tree, and identifies the first vulnerable version, and the last vulnerable versions in the version tree trunk and branches. For evaluation, we took Linux Kernel as the case study and quantified the error rate of the vulnerable versions reported by the NVD. The total number of vulnerable Linux Kernel versions reported by the NVD was 43,727 (as of September 2020), of which the total number of false positives reached 2,497 and the total number of false negatives reached 9,330, accounting for 5.7% and 21.34%, respectively. In addition, we compare our tool with two vulnerability detection tools and show that our tool could achieve high detection accuracy.

Federated recommendation (FR) addresses privacy concerns in recommender systems by training a global model without requiring raw user data to leave individual devices. A server, known as the aggregator, integrates users’ local gradients and updates the global model parameters. However, FR is vulnerable to attacks where malicious users manipulate these updates, known as model poisoning attacks. In this work, we propose a new targeted attack calledStairClimbingto promote specific items through model poisoning, and a new defence mechanismCrossEU. StairClimbingadopts a new strategy resembling stair climbing to enable target items to beat competitive items and increase their popularity level by level. Compared to prior attacks,StairClimbingguarantees balanced effectiveness, efficiency and stealthiness simultaneously. Our defence mechanismCrossEUleverages two patterns regarding the lists of items updated by benign users between iterative epochs. Extensive experiments on six real-world datasets demonstrateStairClimbing’s superiority across all three desirable attack properties, even with a small proportion of malicious users (1%). In addition,CrossEUeffectively delays the impact of all tested attacks and even eliminates their damage entirely.

Recent studies have shown that Deep Neural Networks (DNNs) are easily deceived by adversarial examples, revealing their serious vulnerability. Due to the transferability, adversarial examples can attack across multiple models with different architectures, called transfer-based black-box attacks. Input transformation is one of the most effective methods to improve adversarial transferability. In particular, the attacks fusing other categories of image information reveal the potential direction of adversarial attacks. However, the current techniques rely on input transformations in the spatial domain, which ignore the frequency information of the image and limit its transferability. To tackle this issue, we propose Mixed-Frequency Inputs (MFI) based on a frequency domain perspective. MFI alleviates the overfitting of adversarial examples to the source model by considering high-frequency components from various kinds of images in the process of calculating the gradient. By accumulating these high-frequency components, MFI acquires a more steady gradient direction in each iteration, leading to the discovery of better local maxima and enhancing transferability. Extensive experimental results on the ImageNet-compatible datasets demonstrate that MFI outperforms existing transform-based attacks with a clear margin on both Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), which proves MFI is more suitable for realistic black-box scenarios.

In vehicular ad-hoc networks (VANET), federated learning enables vehicles to collaboratively train a global model for intelligent transportation without sharing their local data. However, due to dynamic network structure and unreliable wireless communication of VANET, various potential risks (e.g., identity privacy leakage, data privacy inference, model integrity compromise, and data manipulation) undermine the trustworthiness of intermediate model parameters necessary for building the global model. While existing cryptography techniques and differential privacy provide provable security paradigms, the practicality of secure federated learning in VANET is hindered in terms of training efficiency and model performance. Therefore, developing a secure and efficient federated learning in VANET remains a challenge. In this work, we propose a privacy-enhanced and efficient authentication protocol for federated learning in VANET, called FedComm. Unlike existing solutions, FedComm addresses the above challenge through user anonymity. First, FedComm enables vehicles to participate in training with unlinkable pseudonyms, ensuring both privacy preservation and efficient collaboration. Second, FedComm incorporates an efficient authentication protocol to guarantee the authenticity and integrity of model parameters originated from anonymous vehicles. Finally, FedComm accurately identifies and completely eliminates malicious vehicles in anonymous communication. Security analysis and verification with ProVerif demonstrate that FedComm enhances privacy and reliability of intermediate model parameters. Experimental results show that FedComm reduces the overhead of proof generation and verification by 67.38% and 67.39%, respectively, compared with the state-of-the-art authentication protocols used in federated learning.

Dynamic link prediction aims to predict future connections among unconnected nodes in a network. It can be applied for friend recommendations, link completion, and other tasks. Network representation learning algorithms have demonstrated considerable effectiveness in various prediction tasks. However, most network representation learning algorithms are based on homogeneous networks and static networks for link prediction that do not consider rich semantic and dynamic information. Additionally, existing dynamic network representation learning methods neglect the neighborhood interaction structure of the node. In this work, we design a neighbor-enhanced dynamic heterogeneous attributed network embedding method (NeiDyHNE) for link prediction. In light of the impressive achievements of the heuristic methods, we learn the information of common neighbors and neighbors’ interaction in heterogeneous networks to preserve the neighbors proximity and common neighbors proximity. NeiDyHNE encodes the attributes and neighborhood structure of nodes as well as the evolutionary features of the dynamic network. More specifically, NeiDyHNE consists of the hierarchical structure attention module and the convolutional temporal attention module. The hierarchical structure attention module captures the rich features and semantic structure of nodes. The convolutional temporal attention module captures the evolutionary features of the network over time in dynamic heterogeneous networks. We evaluate our method and various baseline methods on the dynamic link prediction task. Experimental results demonstrate that our method is superior to baseline methods in terms of accuracy.

Are Federated Learning (FL) systems free from backdoor poisoning with the arsenal of various defense strategies deployed? This is an intriguing problem with significant practical implications regarding the utility of FL services. Despite the recent flourish of poisoning-resilient FL methods, our study shows that carefully tuning the collusion between malicious participants can minimize the trigger-induced bias of the poisoned local model from the poison-free one, which plays the key role in delivering stealthy backdoor attacks and circumventing a wide spectrum of state-of-the-art defense methods in FL. In our work, we instantiate the attack strategy by proposing a distributed backdoor attack method, namely Cerberus Poisoning (CerP). It jointly tunes the backdoor trigger and controls the poisoned model changes on each malicious participant to achieve a stealthy yet successful backdoor attack against a wide spectrum of defensive mechanisms of federated learning techniques. Our extensive study on 3 large-scale benchmark datasets and 13 mainstream defensive mechanisms confirms that Cerberus Poisoning raises a significantly severe threat to the integrity and security of federated learning practices, regardless of the flourish of robust Federated Learning methods.

Delegated-Proof-of-Stake (DPoS) blockchains, such as EOSIO, Steem and TRON, are governed by a committee of block producers elected via a coin-based voting system. We recently witnessed the first de facto blockchain takeover that happened between Steem and TRON. Within one hour of this incident, TRON founder took over the entire Steem committee, forcing the original Steem community to leave the blockchain that they maintained for years. This is a historical event in the evolution of blockchains and Web 3.0. Despite its significant disruptive impact, little is known about how vulnerable DPoS blockchains are in general to takeovers and the ways in which we can improve their resistance to takeovers.

Recent work shows that well-designed adversarial examples can fool deep neural networks (DNNs). Due to their transferability, adversarial examples can also attack target models without extra information, called black-box attacks. However, most existing ensemble attacks depend on numerous substitute models to cover the vulnerable subspace of a target model. In this work, we find three types of models with non-overlapping vulnerable frequency regions, which can cover a large enough vulnerable subspace. Based on this finding, we propose a lightweight ensemble adversarial attack named LEA2, integrated by standard, weakly robust, and robust models. Moreover, we analyze Gaussian noise from the perspective of frequency and find that Gaussian noise is located in the vulnerable frequency regions of standard models. Therefore, we substitute standard models with Gaussian noise to ensure the use of high-frequency vulnerable regions while reducing attack time consumption. Experiments on several image datasets indicate that LEA2achieves better transferability under different defended models compared with extensive baselines and state-of-the-art attacks.

Accurate diagnosis of cancer subtypes is crucial for precise treatment, because different cancer subtypes are involved with different pathology and require different therapies. Although deep learning techniques have made great success in computer vision and other fields, they do not work well on Lung cancer subtype diagnosis, due to the distinction of slide images between different cancer subtypes is ambiguous. Furthermore, they often over-fit to high-dimensional genomics data with limited samples, and do not fuse the image and genomics data in a sensible way. In this paper, we propose a hybrid deep network based approach LungDIG for Lung cancer subtype Diagnosis by fusing Image-Genomics data. LungDIG first tiles the tissue slide image into small patches and extracts the patch-level features by fine-tuning an Inception-V3 model. Since the patches may contain some false positives in non-diagnostic regions, it further designs a patch-level feature combination strategy to integrate the extracted patch features and maintain the diversity between different cancer subtypes. At the same time, it extracts the genomics features from Copy Number Variation data by an attention based nonlinear extractor. Next, it fuses the image and genomics features by an attention based multilayer perceptron (MLP) to diagnose cancer subtype. Experiments on TCGA lung cancer data show that LungDIG can not only achieve higher accuracy for cancer subtype diagnosis than state-of-the-art methods, but also have a high authenticity and good interpretability.

Data reconstruction attack has become an emerging privacy threat to Federal Learning (FL), inspiring a rethinking of FL's ability to protect privacy. While existing data reconstruction attacks have shown some effective performance, prior arts rely on different strong assumptions to guide the reconstruction process. In this work, we propose a novel Conditional Generative Instance Reconstruction Attack (CGIR attack) that drops all these assumptions. Specifically, we propose a batch label inference attack in non-IID FL scenarios, where multiple images can share the same labels. Based on the inferred labels, we conduct a “coarse-to-fine” image reconstruction process that provides a stable and effective data reconstruction. In addition, we equip the generator with a label condition restriction so that the contents and the labels of the reconstructed images are consistent. Our extensive evaluation results on two model architectures and five image datasets show that without the auxiliary assumptions, the CGIR attack outperforms the prior arts, even for complex datasets, deep models, and large batch sizes. Furthermore, we evaluate several existing defense methods. The experimental results suggest that pruning gradients can be used as a strategy to mitigate privacy risks in FL if a model tolerates a slight accuracy loss.

Context-based paring is a promising direction for pairing IoT devices constrained in user interfaces (UIs). However, it takes a proximate distance or a long time for IoT devices to sense highly correlated context with enough entropy. In this work, we present a fast and secure approach, namedMPairing, to pairing multiple commercial off-the-shelf (COTS) IoT devices. This approach is based on the key idea that devices co-located within aphysically-secure boundarycan perceive qualified context under the help of human-in-the-loop (HITL). Specifically, we leverage received-signal-strength (RSS) trajectory data with manually-generated interference in a short period as the shared secret to achieve fast and secure pairing. Subsequently, the real-time RSS trajectory data is utilized to generate random numbers in lieu of pre-shared key (PSK), which makes our scheme more resistant to background attacks. We theoretically prove the security of our pairing scheme and implement it in real-world environments. Our experimental results demonstrate that our scheme can effectively defend against malicious devices by imposing a threshold on the similarity of RSS trajectory data. The experimental results also show that, compared with the traditional context-based pairing that takes up to 24 hours, in our scheme it takes only 10 seconds on average for a legitimate device to pass the similarity checking, which is efficient and robust.

We propose to identify compromised mobile devices from a network administrator’s point of view. Intuitively, inadvertent users (and thus their devices) who download apps through untrustworthy markets are often lured to install malicious apps through in-app advertisements or phishing. We thus hypothesize that devices sharing similar apps would have a similar likelihood of being compromised, resulting in an association between a compromised device and its apps. We propose to leverage such associations to identify unknown compromised devices using the guilt-by-association principle. Admittedly, such associations could be relatively weak as it is hard, if not impossible, for an app to automatically download and install other apps without explicit user initiation. We describe how we can magnify such associations by carefully choosing parameters when applying graph-based inferences. We empirically evaluate the effectiveness of our approach on real datasets provided by a major mobile service provider. Specifically, we show that our approach achieves nearly 98% AUC (area under the ROC curve) and further detects as many as 6 ~ 7 times of new compromised devices not covered by the ground truth by expanding the limited knowledge on known devices. We show that the newly detected devices indeed present undesirable behavior in terms of leaking private information and accessing risky IPs and domains. We further conduct in-depth analysis of the effectiveness of graph inferences to understand the unique structure of the associations between mobile devices and their apps, and its impact on graph inferences, based on which we propose how to choose key parameters.

Maritime ships and ports have become increasingly digital and intelligent. While intelligent maritime transportation systems bring convenience to the maritime industry, ship operation and management are also confronted with network risks. The Internet of Things (IoT) installed in the shipborne network collects and monitors the environmental data of the whole ship. It uses the collected data to make decisions to control the ship. The threat of Local Area Network (LAN) of IoT in ships has become an emerging issue. The DNS rebinding attack is a typical attack, which can bypass firewalls and seriously threaten the marine network in security and privacy of the local IoT. DNS rebinding attacks are difficult to model and detect, due to their sophisticated characteristics. In this work, we define threat models of DNS rebinding attacks and propose an effective method for the detection of and the defense against these attacks. First, we define threat models for DNS rebinding attacks. We employ a Markov chain to model the process of DNS rebinding attacks. With the threat modeling, the attack behaviors are clearly characterized and the most relevant attributes are thus extracted. Second, we propose an effective method for the detection of DNS rebinding attacks in the marine transportation system. The detection method includes the initialization method and the verification method, which manages and verifies access permission of equipment information and the service interface of the IoT in the shipborn network. Finally, we simulate the DNS rebinding attacks on the marine IoT. We analyze and test the security and the performance of the initialization method and the verification method in the simulated environment. The extensive experimental results demonstrate that the IoT in marine networks is vulnerable to DNS rebinding. Our method is effective and efficient to detect and defend against DNS rebinding attacks. It thus secures security and privacy in the local IoT on shipboard.

Graph auto-encoder is considered a framework for unsupervised learning on graph-structured data by representing graphs in a low dimensional space. It has been proved very powerful for graph analytics. In the real world, complex relationships in various entities can be represented by heterogeneous graphs that contain more abundant semantic information than homogeneous graphs. In general, graph auto-encoders based on homogeneous graphs are not applicable to heterogeneous graphs. In addition, little work has been done to evaluate the effect of different semantics on node embedding in heterogeneous graphs for unsupervised graph representation learning. In this work, we propose a novel Heterogeneous Graph Attention Auto-Encoders (HGATE) for unsupervised representation learning on heterogeneous graph-structured data. Based on the consideration of semantic information, our architecture of HGATE reconstructs not only the edges of the heterogeneous graph but also node attributes, through stacked encoder/decoder layers. Hierarchical attention is used to learn the relevance between a node and its meta-path based neighbors, and the relevance among different meta-paths. HGATE is applicable to transductive learning as well as inductive learning. Node classification and link prediction experiments on real-world heterogeneous graph datasets demonstrate the effectiveness of HGATE for both transductive and inductive tasks.

Digital assets involved in smart contracts are on the rise. Security vulnerabilities in smart contracts have resulted in significant losses for the blockchain community. Existing smart contract vulnerability detection techniques have been typically single-purposed and focused only on the source code or opcode of contracts. This paper presents a new smart contract vulnerability detection method, which extracts features from different levels of smart contracts to train machine learning models for effective detection of vulnerabilities. Specifically, we propose to extract 2-gram features from the opcodes of smart contracts and token features from the source code using a pre-trained CodeBERT model, thereby capturing the semantic information of smart contracts at different levels. The 2-gram and token features are separately aggregated and then fused and input into machine-learning models to mine the vulnerability features of contracts. Over 10,266 smart contracts are used to verify the proposed method. Widespread reentrancy, timestamp dependence, and transaction-ordering dependence vulnerabilities are considered. Experiments show the fused features can help significantly improve smart contract vulnerability detection compared to the single-level features. The detection accuracy is as high as 98%, 98% and 94% for the three vulnerabilities, respectively. The average detection time is 0.99 second per contract, indicating the proposed method is suitable for automatic batch detection of vulnerabilities in smart contracts.

Software configuration requires that the user sets appropriate values to specified variables, known as configuration parameters, which potentially affect the behaviors of software system. It is an essential means for software reliability, but how to ensure correct configurations remains a great challenge, especially when a large number of parameter settings are involved. Existing studies on misconfiguration diagnosis treat all configurations independently, ignoring the constraints and correlations among different configurations. In this article, we reveal the phenomenon of multi-misconfigurations and present a tool, MMD, for multi-misconfigurations diagnosis. Specifically, MMD consists of two modules: Correlated Configurations Analysis and Primary Misconfigurations Diagnosis. The former determines the correlation among each pair of configurations by analyzing the control and data flows related to each configuration. The latter is responsible for collecting a list of configurations ranked according to their suspiciousness. Combining the outputs of two modules, MMD is able to assist the user in multi-misconfigurations diagnosis. We evaluate MMD on seven popular Java projects: Randoop, Soot, Synoptic, Hdfs, Hbase, Yarn, and Zookeeper. MMD identifies 510 configuration correlations with a 4.9% false positive rate. Furthermore, it effectively diagnoses 22 multi-misconfigurations collected from StackOverflow, outperforming two state-of-the-art baselines.

Machine Learning-as-a-Service systems (MLaaS) have been largely developed for cybersecurity-critical applications, such as detecting network intrusions and fake news campaigns. Despite effectiveness, their robustness against adversarial attacks is one of the key trust concerns for MLaaS deployment. We are thus motivated to assess the adversarial robustness of the Machine Learning models residing at the core of these securitycritical applications with categorical inputs. Previous research efforts on accessing model robustness against manipulation of categorical inputs are specific to use cases and heavily depend on domain knowledge, or require white-box access to the target ML model. Such limitations prevent the robustness assessment from being as a domain-agnostic service provided to various real-world applications. We propose a provably optimal yet computationally highly efficient adversarial robustness assessment protocol for a wide band of ML-driven cybersecurity-critical applications. We demonstrate the use of the domain-agnostic robustness assessment method with substantial experimental study on fake news detection and intrusion detection problems.

Abstract Centralized machine learning methods require the aggregation of data collected from clients. Due to the awareness of data privacy, however, the aggregation of raw data collected by Internet of Things (IoT) devices is not feasible in many scenarios. Federated learning (FL), a kind of distributed learning framework, can be running on multiple IoT devices. It aims to resolve the issues of privacy leakage by training a model locally on the client-side, other than on the server-side that aggregates all the raw data. However, there are still threats of poisoning attacks in FL. Label flipping attacks, typical data poisoning attacks in FL, aim to poison the global model by sending model updates trained by the data with mismatched labels. The central parameter aggregation server is hard to detect the label flipping attacks due to its inaccessibility to the client in a typical FL system. In this work, we are motivated to prevent label flipping poisoning attacks by observing the changes in model parameters that were trained by different single labels. We propose a novel detection method called average weight of each class in its associated fully connected layer. In this method, we detect label flipping attacks by identifying the differences of classes in the data based on the weight assignments in a fully connected layer of the neural network model and use the statistical algorithm to recognize the malicious clients. We conduct extensive experiments on benchmark data like Fashion-MNIST and Intrusion Detection Evaluation Dataset (CIC-IDS2017). Comprehensive experimental results demonstrated that our method has the detection accuracy over 90% for the identification of the attackers flipping labels.

Summary Compared with traditional voting methods, electronic voting can effectively avoid the phenomenon of fraud for personal gains in various links, it is faster and more accurate in the tallying stage. However, many electronic voting systems have many problems such as inability to verify ballots, easy to be forged, and low computing efficiency. We propose an electronic voting protocol based on homomorphic signcryption and blockchain. The protocol makes the voting process public through blockchain and replaces the traditional trusted third party with the smart contract. It uses the homomorphic encryption algorithm and the homomorphic signcryption algorithm to encrypt and sign the ballot and uses their aggregation properties to perform homomorphic tally on the encrypted votes. This not only reduces the excessive burden on the voters but also improves the voting efficiency. At the same time, it can satisfy the security of electronic voting, and the amount of calculation is small, so it is more convenient and flexible to use in large‐scale voting.

Abstract Empirical attacks on Federated Learning (FL) systems indicate that FL is fraught with numerous attack surfaces throughout the FL execution. These attacks can not only cause models to fail in specific tasks, but also infer private information. While previous surveys have identified the risks, listed the attack methods available in the literature or provided a basic taxonomy to classify them, they mainly focused on the risks in the training phase of FL. In this work, we survey the threats, attacks and defenses to FL throughout the whole process of FL in three phases, including Data and Behavior Auditing Phase , Training Phase and Predicting Phase . We further provide a comprehensive analysis of these threats, attacks and defenses, and summarize their issues and taxonomy. Our work considers security and privacy of FL based on the viewpoint of the execution process of FL. We highlight that establishing a trusted FL requires adequate measures to mitigate security and privacy threats at each phase. Finally, we discuss the limitations of current attacks and defense approaches and provide an outlook on promising future research directions in FL.

Usage-based insurance (UBI) provides reasonable vehicle insurance premiums based on vehicle usage and driving behavior. In general, there are three major issues in realizing intelligent UBI systems. First, UBI evaluation mechanisms are not auditable to drivers. Insurers may thus deliberately adjust the UBI premiums. Second, the process of collecting driving data by insurers may lead to serious privacy breaches. Third, forging safer driving data for reducing insurance premiums may cause economic losses for insurers. To address these challenges, in this study, we propose CCUBI, a cross-chain-based premium competition scheme with privacy preservation for intelligent UBI systems. We introduce tamper-resistant blockchain and smart contracts to construct credible insurance mechanisms. The cross-chain technology connects these blockchains in the entire network to form an open premium competition scheme. Vehicle owners can assess designated insurers by sharing historical data with them to get a suitable CCUBI plan. In addition, we propose a data aggregation method used for CCUBI analysis with privacy preservation. Vehicle owners only publish proofs of the driving data. Proofs can still maintain privacy and computability in cross-chain flows. Finally, we adopt roadside units to detect forged driving data. We conduct a detailed security analysis. Experimental results also demonstrate the efficiency of CCUBI.

Malware detection is a vital task for cybersecurity. For malware dynamic behavior, threats come from a small number of Application Programming Interfaces (APIs) embedded in the API sequences, which are easily ignored or obfuscated in the detection process. Prior works proposed graph-based learning methods to solve this problem using API-level behavior relations. However, the malware detection is still challenging, due to the ignore of the temporal correlation between malicious behaviors. In this study, we model the software behaviors with multiscaled API graph sequences to represent API-level behaviors as well as graph-level temporal behavior correlations. We then propose a novel Dynamic Evolving Graph Convolutional Network (DEGCN) model to capture dynamic evolving pattern of both local API-level and global graph-level software behaviors. In particular, we first extract the API-level (node) representations to capture the directed graph representations for each time slot. We then propose a Graph-encoding-based Gate Recurrent Unit (GGRU) network to capture the graph-level evolving features and their evolving status. The graph features of different time slots and different graph scales are concatenated to detect whether the software is benign or malicious. Our evaluation with two public benchmarks reports that DEGCN achieves the best performance compared with state-of-the-art algorithms.

Since big data becomes a main impetus to the next generation of IT industry, data privacy has received considerable attention in recent years. To deal with the privacy challenges, differential privacy has been widely discussed and related private mechanisms are proposed as privacy-enhancing techniques. However, with today’s differential privacy techniques, it is difficult to generate a sanitized dataset that can suit every machine learning task. In order to adapt to various tasks and budgets, different kinds of privacy mechanisms have to be implemented, which inevitably incur enormous costs for computation and interaction. To this end, in this article, we propose two novel schemes for outsourcing differential privacy. The first scheme efficiently achieves outsourcing differential privacy by using our preprocessing method and secure building blocks. To support the queries from multiple evaluators, we give the second scheme that employs a trusted execution environment to aggregately implement privacy mechanisms on multiple queries. During data publishing, our proposed schemes allow providers to go off-line after uploading their datasets, so that they achieve a low communication cost which is one of the critical requirements for a practical system. Finally, we report an experimental evaluation on UCI datasets, which confirms the effectiveness of our schemes.

User rating of a service is the explicit behavior of users expressing their preference for the service. Most exciting recommendation methods focus on predicting user-service ratings according to users' historical rating behaviors. However, the behavior of users invoking services is implicit feedback. By analyzing the services called by users, mining their potential semantic representations can also help model users' hidden interests. To this end, how to integrate the implicit feedback and explicit rating of users to provide users with better recommendation experience is a problem to be addressed for service recommendation. In this paper, we propose a novel latent semantic integrated explicit rating (LSIER) scheme to recommend services to users. The LSIER scheme is designed by integrating the probabilistic matrix factorization (PMF) model and the probabilistic latent semantic index (PLSI) model. consists of the two stages: (1) the PMF model is used to generate a user feature matrix and a service feature matrix, and the two feature matrices are updated to complete the missing service score records of the users, and (2) the PLSI model is used to train users access records, where an expectation maximization algorithm is applied to derive the model parameters to realize unsupervised soft clustering of services. When the user gives explicit or implicit feedback to the service, the LSIER scheme can identify the current interest probability distribution of the user according to the category to which the called service belongs, and provide the user with a list of service recommendations with scores. The performance of the proposed LSIER scheme is evaluated using the Netflix data set and the Movielens data set. Experiments show that the scheme can achieve better recommendation accuracy and recall rate than existing methods.

With the maturity of Internet-of-Things technology, location-based service (LBS) is developing rapidly in intelligent terminal devices, and it brings new vitality to the fields of logistics, transportation, product traceability and so on. The popularity of LBS produces a lot of spatial data, which inevitably brings burden to the storage and management of LBS provider (LBSP). With the help of cloud computing and cloud storage, outsourcing spatial data to cloud server has become a new trend. However, due to the cloud server is not trusted, data outsourcing will face the problems of data disclosure and query disclosure. Range query is a common query in LBS, considering the situation of data outsourcing, this article proposes an accurate range query (ARQ) scheme, which can realize efficient range query while preserving LBSP's data privacy and user's query privacy from being disclosed to the cloud server. The ARQ scheme is suitable for spatial data in any form without being limited to the case that the data points are only integers, which has a certain practical significance. In addition, by dividing the region into atomic regions, ARQ can realize sublinear search time and ensure dynamic update of spatial data. We proved the security of the proposed scheme through security analysis, and demonstrated the effectiveness of the scheme through experiments.

Socialnet becomes an important component in real life, drawing a lot of study issues of security and safety. Recently, for the features of graph structure in socialnet, adversarial attacks on node classification are exposed, and automatic attack methods such as fast gradient attack (FGA) and NETTACK are developed for per-node attacks, which can be utilized for multinode attacks in a sequential way. However, due to the overlook of perturbation influence between different per-node attacks, the above sequential method does not guarantee a global attack success rate for all target nodes, under a fixed budget of perturbation. In this paper, we propose a parallel adversarial attack framework on node classification. We redesign new loss function and objective function for nonconstraint and constraint perturbations, respectively. Through constructing intersection and supplement mechanisms of perturbations, we then integrate node filtering-based P-FGA and P-NETTACK in a unified framework, finally realizing parallel adversarial attacks. Experiments on politician socialnet dataset Polblogs with detailed analysis are conducted to show the effectiveness of our approach.

Safe and reliable intelligent charging stations are imperative in an intelligent transportation infrastructure. Over the past few years, a big number of smart charging stations have been deployed, and most of them are online and connected, resulting in potential risks of threats. Although there exists related work on securing intelligent vehicles, very little work focused on the security of charging devices. Unlike traditional network systems, these power-related Industrial Control Systems (ICSs) use many different proprietary protocols and diverse interactions. Traditional anomaly detection methods based on network traffic are thus not suitable for these systems. In this work, we propose an anomaly detection method in real vehicle power supply systems based on a deep architecture model. In particular, we propose a novel traffic anomaly detection model based on Multi-Head Attentions (MHA) that take into account the inherent correlations of traffic generated by ICSs. The MHA model is employed to substitute the traditional feature extraction and rule making process with an acceptable computational cost for classifying traffic data. It is an attention-based model that employs Google Transformer encoder architecture to extract recessive features of traffic for anomaly detection. The effectiveness of the model is demonstrated by experiments on two real-world power ICS testbeds including a substation with a slave charging station and a power generation simulation platform based on a distributed control system. Comprehensive experimental results indicate that the MHA model outperforms the Convolutional Neural Networks (CNN)-based and classical machine learning detection models with an accuracy rate of 99.86%.

Understanding the surrounding scenes is one of the fundamental tasks in intelligent transportation systems (ITS), especially in unpredictable driving scenes or in developing regions/cities without digital maps. Street view is the most common scene during driving. Since streets are often full of shops with signboards, scene text recognition over the shop sign images in street views is of great significance and utility to urban scene understanding in ITS. To advance research in this field, (1) we build ShopSign, which is a large-scale scene text dataset of Chinese shop signs in street views. It contains 25,770 natural scene images, and 267,049 text instances. The images in ShopSign were captured in different scenes, from downtown to developing regions, and across 8 provinces and 20 cities in China, using more than 50 different mobile phones. It is very sparse and imbalanced in nature. (2) we carry out a comprehensive empirical study on the performance of state-of-the-art DL based scene text reading algorithms on ShopSign and three other Chinese scene text datasets, which has not been addressed in the literature before. Through comparative analysis, we demonstrate that language has a critical influence on scene text detection. Moreover, by comparing the accuracy of four scene text recognition algorithms, we show that there is a very large room for further improvements in street view text recognition to fit real-world ITS applications.

In blockchain networks, reaching effective supervision while maintaining anonymity to the public has been an ongoing challenge. In existing solutions, certification authorities need to record all pairs of identities and pseudonyms, which is demanding and costly. This article proposed an account guarantee scheme to realize feasible supervision for existing anonymous blockchain networks with lower storage costs. Users are able to guarantee anonymous accounts with account guarantee key pairs generated from certificated polynomial functions, which inherently maintains one-to-n mapping certifications. Single or limited account guarantee key pairs do not leak privacy. Victims are able to request TCs to screen a cheater or disclose a cheater with enough fraud transactions by themselves. Detailed security and privacy analysis showed that the account guarantee scheme preserves user privacy and realizes feasible supervision. Experimental results demonstrated that the account guarantee scheme is efficient and practical.

Android Apps have become the most important mobile applications in the evolving mobile IoT systems, whose security and privacy are confronted with ever more challenges, since such mobile devices as smartphones involve too much personal privacy information. Meanwhile, the developers prefer to put core functions (e.g., encryption function and T9 search function) of Android applications in the native layer for execution efficiency. However, there are no automated security analysis tools to protect the security and privacy of the Android native layer, especially for those dynamically loaded third-party SO libraries. In order to solve the previous problem, which is confusing, we propose a novel and scalable system, called SoProtector, to prevent privacy from leaking via the analysis of data flow between the Java and native layers. For detection of the malicious function implanted in the SO libraries, SoProtector realizes a real-time engine. We derive the malware features via three steps: 1) present binary files in native family as a grayscale image; 2) with use of the ARM instructions set reversely obtain the code of the SO file and using Python to obtain the opcode sequence; and 3) each file is transformed as the form of assembly language by IDA Pro, which includes a gdl file as an accompaniment. Our experiment, which involved 3400 applications, demonstrates that SoProtector is able to detect more sinks, sources, and smudges. It effectively inspects and blocks at least 82% of the applications that are loading malicious third-party SO dynamically, and it has relatively low overhead in the meantime, compared to most of the existing static analysis tools (e.g., FlowDroid and AndroidLeaks).

Social networks have integrated into the daily lives of most people in the way of interactions and of lifestyles. The users' identity, relationships, or other characteristics can be explored from the social networking data, in order to provide personalized services to the users. In this article, we focus on predicting the user's emotional intelligence (EI) based on social networking data. As an essential facet of users' psychological characteristics, EI plays an important role on well-being, interpersonal relationships, and overall success in people's life. Perception of EI contributes to predicting one's behavior or group behavior. Most existing work on predicting people's EI is based on questionnaires that may collect dishonest answers or unconscientious responses, thus leading in potentially inaccurate prediction results. In this article, we are motivated to propose EI prediction models based on the sentiment analysis of social networking data. The models are represented by four dimensions, including self-awareness, self-regulation, self-motivation, and social relationships. The EI of a user is then measured by four numerical values or the sum of them. In the experiments, we predict the EIs of over a hundred thousand users based on one of the largest social networks of China, Weibo. The predicting results demonstrate the effectiveness of our models. The results show that the distribution of the four EI's dimensions of users is roughly normal. The results also indicate that EI scores of females are generally higher than males' EI scores. This is consistent with previous findings. In addition, the four dimensions of EI are correlated. We finally analyze the advantages and the disadvantages of our models in predicting users' EI with social networking data.

Millions of crimes arise each year, which threatens public safety and harms the victims. Precise crime inference is of great significance in preventing crimes. The sharply increasing large-scale heterogeneous data provide a chance to reveal the patterns and trends in crimes. Several approaches employing feature-based regression or spatiotemporal distribution fitting are proposed but lack of some considerations: 1) ignore the dynamic mutual influences among crimes and locations; and 2) overlook the large scale, incompleteness, uncertainty, and vagueness in the heterogeneous data. This article comprehensively investigates the reliability and applicability of proposing a coevolutionary model to formulate the interaction pattern among the crimes and locations and develops a fuzzy-rough-set-based feature selection method to discover the distinctiveness and permanence properties of the crimes and locations with different latent features. Extensive experiments show that our algorithm achieves the mean absolute error of 1.529 (hour) in the crime time inference and the accuracy of 0.653 and 0.633 in the crime type and location inference, which surpass the state of the arts more than 6.5, 1.9, and 1.8 times, respectively. Additional experiments on different parameter settings of our model are provided to further explore its effectiveness and scalability.

While much effort has been made to detect and measure the privacy leakage caused by the advertising (ad) libraries integrated in mobile applications, analytics libraries, which are also widely used in mobile apps have not been systematically studied for their privacy risks. Different from ad libraries, the main function of analytics libraries is to collect users' in-app actions. Hence, by design analytics libraries are more likely to leak users' private information. In this work, we study what information is collected by the analytics libraries integrated in popular Android apps. We design and implement a framework called “Alde”. Given an app, Alde employs both static analysis and dynamic analysis to detect the users' in-app actions collected by analytics libraries. We also study what private information can be leaked by the apps that use the same analytics library. Moreover, we analyze apps' privacy policies to see whether app developers have notified the users that their in-app action data is collected by analytics libraries. Finally, we select eight widely used analytics libraries to study and apply our method to 300 popular apps downloaded from both Chinese app markets and Google play. Our experimental results show that some apps indeed leak users' personal information through analytics libraries even though their genuine purposes of using analytics services are legal. To mitigate such threats, we have developed an app named “ALManager” that leverages the Xposed framework to manage analytics libraries in other apps.

Assisting traffic control is one of the most important applications on the Internet of Vehicles (IoVs). Traffic information provided by vehicles is desired since drivers or vehicle sensors are sensitive in perceiving or detecting nuances on roads. However, the availability and privacy preservation of this information are critical while conflicted with each other in the vehicular communication. In this paper, we propose a semicentralized mode with attribute-based blockchain in IoVs to balance the tradeoff between the availability and the privacy preservation. In this mode, a method of control-by-vehicles is used to control signals of traffic lights to increase traffic efficiency. Users are grouped their attributes such as locations and directions before starting the communication. The users reach an agreement on determining a temporary signal timing by interacting with each other without leaking privacy. Final decisions are verifiable to all users, even if they have no a priori agreement and processes of consensus. The mode not only achieves the aim of privacy preservation but also supports responsibility investigation for historical agreements via ciphertext-policy attribute-based encryption (CP-ABE) and blockchain technology. Extensive experimental results demonstrated that our mode is efficient and practical.

In recent years, wireless sensor networks (WSNs) have become an active area of research for monitoring physical and environmental conditions. Due to the interdependence of sensors, a functional anomaly in one sensor can cause a functional anomaly in another sensor, which can further lead to the malfunctioning of the entire sensor network. Existing research work has analysed faulty sensor anomalies but fails to show the effectiveness throughout the entire interdependent network system. In this article, a dictionary learning algorithm based on a non-negative constraint is developed, and a sparse representation anomaly node detection method for sensor networks is proposed based on the dictionary learning. Through experiment on a specific thermal power plant in China, we verify the robustness of our proposed method in detecting abnormal nodes against four state of the art approaches and proved our method is more robust. Furthermore, the experiments are conducted on the obtained abnormal nodes to prove the interdependence of multi-layer sensor networks and reveal the conditions and causes of a system crash.

With wide use of cloud computing and storage services, sensitive information is increasingly centralized into the cloud to reduce the management costs, which raises concerns about data privacy. Encryption is a promising way to maintain the confidentiality of outsourced sensitive data, but it makes effective data utilization to be a very challenging task. In this paper, we focus on the problem of private matching over outsourced encrypted datasets in identity-based cryptosystem that can simplify the certificate management. To solve this problem, we propose an Identity-Based Private Matching scheme (IBPM), which realizes fine-grained authorization that enables the privileged cloud server to perform private matching operations without leaking any private data. We present the rigorous security proof under the Decisional Linear Assumption and Decisional Bilinear Diffie-Hellman Assumption. Furthermore, through the analysis of the asymptotic complexity and the experimental evaluation, we verify that the cost of our IBPM scheme is linear to the size of the dataset and it is more efficient than the existing work of Zheng and Xu [30]. Finally, we apply our IBPM scheme to build two efficient schemes, including identity-based fuzzy private matching as well as identity-based multi-keyword fuzzy search.

The vehicular announcement network is one of the most promising utilities in the communications of smart vehicles and in the smart transportation systems. In general, there are two major issues in building an effective vehicular announcement network. First, it is difficult to forward reliable announcements without revealing users' identities. Second, users usually lack the motivation to forward announcements. In this paper, we endeavor to resolve these two issues through proposing an effective announcement network called CreditCoin, a novel privacy-preserving incentive announcement network based on Blockchain via an efficient anonymous vehicular announcement aggregation protocol. On the one hand, CreditCoin allows nondeterministic different signers (i.e., users) to generate the signatures and to send announcements anonymously in the nonfully trusted environment. On the other hand, with Blockchain, CreditCoin motivates users with incentives to share traffic information. In addition, transactions and account information in CreditCoin are tamper-resistant. CreditCoin also achieves conditional privacy since Trace manager in CreditCoin traces malicious users' identities in anonymous announcements with related transactions. CreditCoin thus is able to motivate users to forward announcements anonymously and reliably. Extensive experimental results show that CreditCoin is efficient and practical in simulations of smart transportation.

MOTIVATION: Growth phenotype profiling of genome-wide gene-deletion strains over stress conditions can offer a clear picture that the essentiality of genes depends on environmental conditions. Systematically identifying groups of genes from such high-throughput data that share similar patterns of conditional essentiality and dispensability under various environmental conditions can elucidate how genetic interactions of the growth phenotype are regulated in response to the environment. RESULTS: We first demonstrate that detecting such 'co-fit' gene groups can be cast as a less well-studied problem in biclustering, i.e. constant-column biclustering. Despite significant advances in biclustering techniques, very few were designed for mining in growth phenotype data. Here, we propose Gracob, a novel, efficient graph-based method that casts and solves the constant-column biclustering problem as a maximal clique finding problem in a multipartite graph. We compared Gracob with a large collection of widely used biclustering methods that cover different types of algorithms designed to detect different types of biclusters. Gracob showed superior performance on finding co-fit genes over all the existing methods on both a variety of synthetic data sets with a wide range of settings, and three real growth phenotype datasets for E. coli, proteobacteria and yeast. AVAILABILITY AND IMPLEMENTATION: Our program is freely available for download at http://sfb.kaust.edu.sa/Pages/Software.aspx. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

With the exponential growth of smartphone adoption, malware attacks on smartphones have resulted in serious threats to users, especially those on popular platforms, such as Android. Most Android malware is generated by piggybacking malicious payloads into benign applications (apps), which are called piggybacked apps. In this paper, we propose DAPASA, an approach to detect Android piggybacked apps through sensitive subgraph analysis. Two assumptions are established to reflect the different invocation patterns of sensitive APIs in the injected malicious payloads (rider) of a piggybacked app and in its host app (carrier). With these two assumptions, DAPASA generates a sensitive subgraph (SSG) to profile the most suspicious behavior of an app. Five features are constructed from SSG to depict the invocation patterns. The five features are fed into the machine learning algorithms to detect whether the app is piggybacked or benign. DAPASA is evaluated on a large real-world data set consisting of 2551 piggybacked apps and 44 921 popular benign apps. Extensive evaluation results demonstrate that the proposed approach exhibits an impressive detection performance compared with that of three baseline approaches even with only five numeric features. Furthermore, the proposed approach can complement permission-based approaches and API-based approaches with the combination of our five features from a new perspective of the invocation structure.

Smartphone users are facing serious threat of privacy leakage. This privacy leakage is caused not only by malicious applications (apps), but also by the most popular apps in one's pocket. In this poster, we present our study on the issues of information leakage caused by the most widely used apps in Chinese app markets. Our goal is to find what information is exposed by each popular app, and then to focus on the following three questions in order to explore the influence of this kind of information leakage: (1) to what extent can the information leaked by an app be used to characterize the user's behaviors? (2) to what extent can the information leaked by a number of apps in the same smartphone be used to characterize the user's behaviors' and (3) whether the leaked information from a number of smartphones can be integrated to predict the social behaviors' Preliminary experimental results on the top 50 popular apps in Chinese app markets show the serious situation of this kind of information leakage.

Today's Android-powered smartphones are equipped with various embedded sensors, such as the motion sensors, the environmental sensors and the position sensors. Many functions in the third-party applications (apps) need to use these sensors. However, embedded sensors may lead to security issues, as the third-party apps can access data from these sensors without claiming any permissions. It has been proven that embedded sensors can be exploited by well designed malicious apps, resulting in leaking users' privacy. In this work, we are motivated to provide an up-to-date overview of sensor usage patterns in current apps by investigating what, why and how embedded sensors are used in all the apps collected from a complete app market. To fulfill this goal, We develop a tool called “SDFDroid” to identify the sensors' types and to generate the sensor data propagation paths in each app. We then cluster the apps to find out their sensor usage patterns based on their sensor data propagation paths. We apply our method on AppChina, a widely used Chinese Android app market. Extensive experiments are conducted and the experimental results show that most apps implement their sensor related functions by using the third-party libraries. We further study the sensor usage in the third-party libraries. Our results show the accelerometer sensor is the most frequently used sensor. Though many third-party libraries use no more than four types of sensors, there are some third-party libraries register all the types of sensors recklessly. These results show the need for better regulating the sensor usage in Android apps.

Efficiently processing massive data is a big issue in high‐speed network intrusion detection, as network traffic has become increasingly large and complex. In this work, instead of constructing a large number of features from massive network traffic, the authors aim to select the most important features and use them to detect intrusions in a fast and effective manner. The authors first employed several techniques, that is, information gain (IG), wrapper with Bayesian networks (BN) and Decision trees (C4.5), to select important subsets of features for network intrusion detection based on KDD'99 data. The authors then validate the feature selection schemes in a real network test bed to detect distributed denial‐of‐service attacks. The feature selection schemes are extensively evaluated based on the two data sets. The empirical results demonstrate that with only the most important 10 features selected from all the original 41 features, the attack detection accuracy almost remains the same or even becomes better based on both BN and C4.5 classifiers. Constructing fewer features can also improve the efficiency of network intrusion detection.

Android has been a major target of malicious applications (malapps). How to detect and keep the malapps out of the app markets is an ongoing challenge. One of the central design points of Android security mechanism is permission control that restricts the access of apps to core facilities of devices. However, it imparts a significant responsibility to the app developers with regard to accurately specifying the requested permissions and to the users with regard to fully understanding the risk of granting certain combinations of permissions. Android permissions requested by an app depict the app's behavioral patterns. In order to help understanding Android permissions, in this paper, we explore the permission-induced risk in Android apps on three levels in a systematic manner. First, we thoroughly analyze the risk of an individual permission and the risk of a group of collaborative permissions. We employ three feature ranking methods, namely, mutual information, correlation coefficient, and T-test to rank Android individual permissions with respect to their risk. We then use sequential forward selection as well as principal component analysis to identify risky permission subsets. Second, we evaluate the usefulness of risky permissions for malapp detection with support vector machine, decision trees, as well as random forest. Third, we in depth analyze the detection results and discuss the feasibility as well as the limitations of malapp detection based on permission requests. We evaluate our methods on a very large official app set consisting of 310 926 benign apps and 4868 real-world malapps and on a third-party app sets. The empirical results show that our malapp detectors built on risky permissions give satisfied performance (a detection rate as 94.62% with a false positive rate as 0.6%), catch the malapps' essential patterns on violating permission access regulations, and are universally applicable to unknown malapps (detection rate as 74.03%).

The Affinity Propagation (AP) clustering algorithm proposed by Frey and Dueck (2007) provides an understandable, nearly optimal summary of a data set. However, it suffers two major shortcomings: i) the number of clusters is vague with the user-defined parameter called self-confidence, and ii) the quadratic computational complexity. When aiming at a given number of clusters due to prior knowledge, AP has to be launched many times until an appropriate setting of self-confidence is found. The re-launched AP increases the computational cost by one order of magnitude. In this paper, we propose an algorithm, called K-AP, to exploit the immediate results of K clusters by introducing a constraint in the process of message passing. Through theoretical analysis and experimental validation, K-AP was shown to be able to directly generate K clusters as user defined, with a negligible increase of computational cost compared to AP. In the meanwhile, K-AP preserves the clustering quality as AP in terms of the distortion. K-AP is more effective than k-medoids w.r.t. the distortion minimization and higher clustering purity.

Detection of web attacks is an important issue in current defense-in-depth security framework. In this paper, we propose a novel general framework for adaptive and online detection of web attacks. The general framework can be based on any online clustering methods. A detection model based on the framework is able to learn online and deal with "concept drift" in web audit data streams. Str-DBSCAN that we extended DBSCAN to streaming data as well as StrAP are both used to validate the framework. The detection model based on the framework automatically labels the web audit data and adapts to normal behavior changes while identifies attacks through dynamical clustering of the streaming data. A very large size of real HTTP Log data collected in our institute is used to validate the framework and the model. The preliminary testing results demonstrated its effectiveness.

DDoS attacks are major threats in current computer networks. However, DDoS attacks are difficult to be quickly detected. In this paper, we introduce a system that only extracts several important attributes from network traffic for DDoS attack detection in real computer networks. We collect a large set of DDoS attack traffic by implementing various DDoS attacks as well as normal data during normal usage. Information Gain and Chi-square methods are used to rank the importance of 41 attributes extracted from the network traffic with our programs. Bayesian networks as well as C4.5 are then employed to detect attacks as well as to determine what size of attributes is appropriate for fast detection. Empirical results show that only using the most important 9 attributes, the detection accuracy remains the same or even has some improvements compared with that of using all the 41 attributes based on Bayesian Networks and C4.5 methods. Only using several attributes also improves the efficiency in terms of attributes constructing, models training as well as intrusion detection.

Most current anomaly intrusion detection systems (IDSs) detect computer network behavior as normal or abnormal but cannot identify the type of attacks. Moreover, most current intrusion detection methods cannot process large amounts of audit data for real-time operation. In this paper, we propose a novel method for intrusion identification in computer networks based on principal component analysis (PCA). Each network connection is transformed into an input data vector. PCA is employed to reduce the dimensionality of the data vectors and identification is handled in a low dimensional space with high efficiency and low use of system resources. The normal behavior is profiled based on normal data for anomaly detection and models of each type of attack are built based on attack data for intrusion identification. The distance between a vector and its reconstruction onto those reduced subspaces representing the different types of attacks and normal activities is used for identification. The method is tested with network data from MIT Lincoln labs for the 1998 DARPA intrusion detection evaluation program and testing results show that the model is promising in terms of identification accuracy and computational efficiency for real-time intrusion identification.