Blockchain phishing scams cause billions in annual losses, yet extreme data imbalance severely hampers existing detection algorithms. Current dynamic graph generation methods fragment structures and generate erroneous connections, failing to capture local dynamic patterns vital for node classification. This raises critical questions: Can models minimize isolated subgraph generation? How can they learn and replicate structured, recurring interaction patterns? To answer these questions, we introduce GraphFlowGen, an end-to-end deep generative framework. To minimize isolated subgraph generation, GraphFlowGen employs a novel preprocessing module that jointly extracts structural and temporal contexts from transaction data, preventing fragmentation and information loss. To learn and replicate structured interaction patterns, it incorporates a Transformer encoder with Graph Attention Networks (GAT) to capture node connection dynamics and temporal evolution. To ensure high fidelity while reducing erroneous links, a reinforcement learning (RL) mechanism iteratively refines generated graph structures. Empirical validation on three real-world datasets demonstrates the effectiveness of our algorithm in local dynamic graph generation and its utility for downstream phishing detection tasks.

Machine learning models are often vulnerable to inference attacks that expose sensitive information from their training data. Shadow model technique is commonly employed in such attacks, like membership inference. However, the need for a large number of shadow models leads to high computational costs, limiting their practical applicability. Such inefficiency mainly stems from the independent training and use of these shadow models. To address this issue, we present a novel shadow pool training framework SHAPOOL, which constructs multiple shared models and trains them jointly within a single process. In particular, we leverage the Mixture-of-Experts mechanism as the shadow pool to interconnect individual models, enabling them to share some sub-networks and thereby improving efficiency. To ensure the shared models closely resemble independent models and serve as effective substitutes, we introduce three novel modules: path-choice routing, pathway regularization, and pathway alignment. These modules guarantee random data allocation for pathway learning, promote diversity among shared models, and maintain consistency with target models. We evaluate SHAPOOL in the context of various membership inference attacks and show that it significantly reduces the computational cost of shadow model construction while maintaining comparable attack performance.

Information leakage issues in machine learning-based Web applications have attracted increasing attention. While the risk of data privacy leakage has been rigorously analyzed, the theory of model function leakage, known as Model Extraction Attacks (MEAs), has not been well studied. In this paper, we are the first to understand MEAs theoretically from an attack-agnostic perspective and to propose analytical metrics for evaluating model extraction risks. By using the Neural Tangent Kernel (NTK) theory, we formulate the linearized MEA as a regularized kernel classification problem and then derive the fidelity gap and generalization error bounds of the attack performance. Based on these theoretical analyses, we propose a new theoretical metric called Model Recovery Complexity (MRC), which measures the distance of weight changes between the victim and surrogate models to quantify risk. Additionally, we find that victim model accuracy, which shows a strong positive correlation with model extraction risk, can serve as an empirical metric. By integrating these two metrics, we propose a framework, namely Model Extraction Risk Inspector (MER-Inspector), to compare the extraction risks of models under different model architectures by utilizing relative metric values. We conduct extensive experiments on 16 model architectures and 5 datasets. The experimental results demonstrate that the proposed metrics have a high correlation with model extraction risks, and MER-Inspector can accurately compare the extraction risks of any two models with up to 89.58%.

Recent studies show that privacy leakages may occur in vertical federated learning (VFL), where parties hold split features of the same samples. While various attacks, including label and feature inference, focus on record-level privacy risks in VFL, few studies delve into the distribution-level privacy threat. In this paper, we explore property inference attacks (PIAs) in VFL, where an adversarial party seeks to deduce global distribution information about a target property in the victim party’s training set. Our key observation is that theLp-norm distribution of intermediate results in VFL could reflect the fraction of the target property in a training set. Inspired by this, we presentProVFL, a novel PIA framework involving distribution comparison and correlation augmentation modules. To achieve property inference, we design a distribution comparison module by creating various intermediate-result populations with different proportions, aiming to learn the relationship betweenLp-norm distributions and their fractions. Then, we theoretically analyze the factors that contribute to the attack effectiveness and develop a correlation augmentation module based on label replacement and model refinement to amplify property information leakage. Extensive experimental results demonstrate that our attacks can achieve inferences with low estimation errors as low as 1%. This poses the immediate threat of property information leakage from private training data in the VFL setting.

The use of log files in digital forensics highlights the importance of ensuring their data integrity for auditing purposes. However, traditional centralized audit log systems face challenges in maintaining data integrity due to log injection attacks and single-point failures. Although blockchain technology can accurately process and replicate log files, existing blockchain-based audit log systems still suffer from security and reliability issues due to their weak threat models and limited scalability. To address these concerns, we propose a blockchain-based audit log system that ensures data integrity under a general threat model where a part of the nodes, including loggers and auditors, are untrusted. First, our proposed system resists collusion attacks by incorporating multiple nodes for system processes and utilizing smart contracts to enforce consensus algorithms. Second, to save blockchain storage space, we design an efficient log integrity proof method, which generates a sub-Non-Fungible Token (sub-NFT) for each log file and keeps it on the blockchain as integrity proof. The single-point failure problem is resolved by outsourcing log files to a distributed file system. To evaluate the proposed system, we implement a prototype based on Hyperledger Fabric. Experimental results show that our proof generation method can reduce storage space usage in comparison to other blockchain-based audit log systems, saving approximately 50% of space in Hyperledger Fabric. The security analysis proves that our system can ensure log file data integrity under the proposed threat model.

Blockchain technology can construct a distributed and trusted ledger, which can be used for electronic voting (E-voting) systems to ensure the security of voting data and improve government credibility. However, existing blockchain-based solutions cannot fully fulfill five core requirements in E-voting, i.e., auditability, privacy, authentication, correctness, and unreusability, which make them unpractical in the reality. In this paper, we propose a Double Blockchain-based E-voting (DBE-voting) system, which consists of a private blockchain and a public blockchain. In the proposed system, the voter information is only recorded in the private blockchain for further auditing and the voting results are recorded in both blockchains. The voter's privacy can be protected in the private blockchain while the voting results can be queried in the public blockchain for verifying the correctness of the election process. Moreover, the ballot recorded in both blockchains is signed with a valid linkable ring signature to ensure authentication and unreusability. We propose an on-chain and off-chain hybrid storage mechanism to ensure the consistency and correctness of voting data in two blockchains. Experimental results demonstrate that the throughput of our system can reach 29 transactions per second when the block size is 512 KB. The security analysis shows that the DBE-voting is the first blockchain-based system that can meet all five requirements simultaneously.

Physical-layer key generation (PKG) establishes cryptographic keys from highly correlated measurements of wireless channels, which relies on reciprocal channel characteristics between uplink and downlink, is a promising wireless security technique for Internet of Things (IoT). However, it is challenging to extract common features in frequency-division duplexing (FDD) systems as uplink and downlink transmissions operate at different frequency bands whose channel frequency responses are not reciprocal anymore. Existing PKG methods for FDD systems have many limitations, i.e., high overhead and security problems. This article proposes a novel PKG scheme that uses the feature mapping function between different frequency bands obtained by deep learning to make two users generate highly similar channel features in FDD systems. In particular, this is the first time to apply deep learning for PKG in FDD systems. We first prove the existence of the band feature mapping function for a given environment and a feedforward network with a single hidden layer can approximate the mapping function. Then, a key generation neural network (KGNet) is proposed for reciprocal channel feature construction, and a key generation scheme based on the KGNet is also proposed. Numerical results verify the excellent performance of the KGNet-based key generation scheme in terms of randomness, key generation ratio, and key error rate. Besides, the overhead analysis shows that the method proposed in this article can be used for resource-constrained IoT devices in FDD systems.

Secret key generation based on wireless channel reciprocity has received widespread attention. However, in frequency division duplexing (FDD) systems, since the carrier frequencies of the uplink and downlink are different and the channel coefficients are no longer reciprocal, key generation for FDD systems is challenging. In this paper, a Complex-Valued neural Network (CVNet) is proposed to predict the downlink channel and generate reciprocal channel characteristics. Then, based on the trained CVNet, we propose a key generation protocol for FDD systems. Numerical results show that the CVNet achieves better performance in terms of prediction accuracy, bit disagreement rate, and bit generation ratio than a traditional Real-Valued Network (RVNet) under high signal-to-noise ratios. Furthermore, the training parameters required by the CVNet account for only half of that required by the RVNet.