Greybox fuzzing has become an effective technique for uncovering vulnerabilities in network protocol implementations. However, existing approaches still face several significant challenges: (1) state modeling is overly coarse-grained, failing to accurately capture subtle state transitions during protocol execution, (2) focusing solely on the first mutation point that reaches the target state, overlooking other regions that may equally impact the target state, (3) neglecting the non-uniform contribution of different message regions to path coverage. To address these issues, we propose UFG-AFLNET, a unified greybox fuzzing framework. UFG-AFLNET significantly enhances fuzzing efficiency and vulnerability discovery through fine-grained state machine modeling, gradient-guided sequence selection, and lightweight dynamic taint inference. Specifically, UFG-AFLNET introduces a state clustering learner that uses the Single-Pass clustering algorithm to extend response state machines into fine-grained path state machines, thereby enabling more precise state differentiation. Additionally, to select the most critical mutation points, we employ a recurrent neural network to compute the sensitivity gradients between target path states and message regions. Finally, we use a message sequence mutator supported by dynamic taint inference to assign weights to each byte and prioritize mutations on those most likely to expose new execution paths. Experiments on five widely used protocol implementations show that UFG-AFLNET significantly outperforms baseline fuzzers in path coverage, the number of vulnerabilities discovered and so on. These results demonstrate the potential of UFG-AFLNET in advancing the field of network protocol security testing.

Anomaly detection on attributed graphs has applications in various domains such as finance and email spam detection, thus gaining substantial attention. Distributed scenarios can also involve issues related to anomaly detection in attribute graphs, such as in medical scenarios. However, most of the existing anomaly detection methods are designed for centralized scenarios, and directly applying them to distributed settings may lead to reduced performance. One possible reason for this issue is that, when graph data are distributed across multiple clients, federated graph learning may struggle to fully exploit the potential of the dispersed data, leading to suboptimal performance. Building on this insight, we propose FedCLGN, a federated graph anomaly detection framework that leverages contrastive self-supervised learning. First, we put forward an augmentation method to maintain global negative pairs on the server. This involves identifying anomalous nodes using pseudo-labels, extracting embedding representations of the negative pairs corresponding to these anomalous nodes from clients, and uploading them to the server. Then, we adopt graph diffusion to enhance the feature representation of nodes, capturing the global structure and local connection patterns. This strategy can strengthen the differentiation between positive and negative instance pairs. Finally, the effectiveness of our approach is verified by experimental results on four real graph datasets.

Image recognition plays a critical role in urban security, traffic management, and environmental monitoring, yet achieving high accuracy in obstructed scenes remains a challenge. To address this, we propose a Graph Convolutional Visual Prior-Transform Network (GCVPN), which significantly improves recognition accuracy and efficiency in complex environments. GCVPN introduces an image prior slicing and topology transformer to convert image data into graph-structured slice features, integrating domain overlap sampling and planar mapping to handle symmetry and enable precise, rapid anomaly detection. By combining a traditional VGG backbone with graph convolutional layers, GCVPN jointly captures topological relationships and feature semantics, while maintaining real-time efficiency with continuous recognition at 30 video frames per second. Extensive experiments demonstrate its effectiveness in photovoltaic panel anomaly detection and face occlusion recognition, highlighting strong potential for applications in intelligent surveillance and autonomous driving.

Anomaly detection in solar photovoltaic panels using Unmanned Aerial Vehicles (UAVs) faces challenges due to minimal texture variations from surface anomalies (e.g., shadows, eddy currents) in UAV-captured imagery, which constrain both detection precision and real-time performance. Existing approaches often lack quantitative anomaly analysis that integrates aerial imagery with operational data, thereby limiting their practical value and impeding sustainable industry advancement. To address these limitations, we propose Gate-convolution Support Vector Data Description (Gate-conv SVDD), a novel framework that enhances the efficiency and accuracy of anomaly detection through rapid parallel gated feature compression and hypersphere-based Support Vector Data Description (SVDD) clustering. This approach enables precise anomaly localization in high-resolution UAV imagery, as validated through simulations and controlled experimental datasets. Gate-conv SVDD further supports quantitative assessments of anomaly severity, thereby bridging the gap between image-based detection and actionable analysis. Designed for computational efficiency, the framework demonstrates strong potential for fast inference in support of real-time UAV imaging, subject to further hardware integration and field validation. Extensive experiments demonstrate that Gate-conv SVDD outperforms state-of-the-art methods, offering superior accuracy and robustness in controlled settings.

Fraud in e-commerce fields (e.g., Amazon, Taobao, and so on) and social networks (e.g., Twitter and Weibo) has recently brought a very bad user experience. Rating fraud detection is an urgent issue for improving user experiences. However, existing methods have lots of limitations in some respects, because it is always very hard to acquire sufficient labeled data for fraud detection and detect new fraud patterns. Fortunately, the relationship for users rating (e.g., purchasing and following) products can be represented as a bipartite graph. So the problem of rating fraud detection can be transformed into the problem of abnormal subgraph detection in the bipartite graph. The major challenge of fraud detection is to distinguish fake rates from real user rates. In this article, we focus on mining rating fraud-connected subgraphs in a bipartite graph. The motivation for this work is fraud detection tasks, which can usually be formulated as mining a bipartite graph formed by source nodes (followers and users) and target nodes (followees and products) for malicious patterns. Now, smart fraudsters evade existing detection methods by buying a large pool of users and hijacking honest users, making them look “normal”-this behavior is called “camouflage.” Accordingly, we propose a fraud detection approach for mining rating fraud subgraph (MRFS), which addresses the problem from the intrinsic metric (e.g., fraudulence, badness and unreliability). The proposed MRFS mines the intrinsic characteristics of nodes and edges from node behavior information, which is an effective and scalable (linear on the input size) algorithm. A large number of comparative experimental results on real-world rating networks show that our proposed MRFS is efficient and universal.

Anomaly detection in multilayer graphs becomes more critical in many application scenarios, i.e., identifying crime hotspots in urban areas by discovering suspicious and illicit behaviors in social networks. However, it is a big challenge to identify anomalies in a layer graph due to the insufficient anomaly features. Most existing methods of anomaly detection determine whether a node is abnormal by looking at the observable anomalous feature values. However, these methods are not suitable for scenarios in which the abnormal features are scarce, e.g., geometric graphs or non-public data in social network services. In this paper, to detect anomaly in a graph with insufficient anomalous features, we propose a pioneering approach ASD-FT (Anomaly Subgraph Detection with Feature Transfer) based on a strategy of anomalous feature transfers between different layers of a multilayer graph. The proposed ASD-FT detects anomaly subgraphs from the graph of the target layer by analyzing the anomalous features in the graph of another layer. We demonstrate the effectiveness and robustness of our approach ASD-FT with extensive experiments on five real-world datasets.