Hypergraph contrastive learning has emerged as a powerful unsupervised paradigm for hypergraph representation learning. Traditional hypergraph contrastive learning methods typically leverage neighbor aggregation strategy to obtain entity (node and hyperedge) representations within each connected component, and then utilize contrastive losses (e.g., node- or hyperedge-level) to update the encoders. However, since entities are usually focused equally on their respective losses, large connected components with numerous entities tend to provide a dominant contribution to the whole learning process, which inevitably hinders the effective learning of entity representations within small connected components. To address this issue, we propose a novel Connected-Component-Aware Hypergraph Contrastive Learning method (CCAHCL). Different from previous methods that only construct node or hyperedge representations, our method additionally constructs the connected component representations, and accordingly designs a hierarchical contrastive loss to balance the model's focus on different scales of connected components. Specifically, we first use the traditional neighbor aggregation strategy to aggregate and update entity (node and hyperedge) representations. Then, these entity representations are further aggregated to generate the connected component representations, where entity features are incorporated into connected components and their structural information is propagated back to enrich their corresponding entities. Afterwards, we employ node-level and hyperedge-level losses to learn the enriched entity representations, and further propose a novel connected-component-level contrastive loss to balance the model's focus on all different connected components, naturally avoiding the learning bias on large connected components. Extensive experiments on various datasets demonstrate that our proposed model achieves superior performance against other state-of-the-art methods.

In multi-view multi-label (MVML) classification, each sample is represented by multiple heterogeneous views and annotated with multiple labels. Existing methods typically exploit pairwise semantic relationships to mine intra-view correlations and align inter-view features for generating structural representations. However, these methods ignore the direct expression of high-order semantic similarities and alignments from a group perspective, which necessitates multi-step aggregation for subsequent feature fusion, leading to the inefficient and incomplete integration of key semantic information. To overcome this limitation, we propose a novel hypergraph-based MVML method with Adaptive High-Order Semantic Fusion (HyperAHSF), which leverages hypergraphs to adaptively model group-level semantic similarities within each view and group-level semantic alignments across different views, enabling more effective feature fusion. Specifically, we first construct view-specific hyperedges by selecting multiple groups of node representations exhibiting high semantic similarity, which captures the group-level semantic similarities within each view, forming view-specific hypergraphs. Furthermore, we establish cross-view hyperedges to connect the multi-view node representations of each sample, which characterizes the group-level semantic alignments across different views, accordingly forming a unified multi-view hypergraph. Afterwards, we employ hypergraph neural networks to efficiently aggregate view-specific information and consensus information from their corresponding hypergraphs via group-level message passing. During the passing process, we impose a label-driven contrastive loss on the consensus information to encourage these representations to cluster toward their corresponding class prototypes, enhancing their discriminability. Finally, the consensus information together with the view-specific information is jointly integrated for multi-label classification. Extensive experiments demonstrate that HyperAHSF outperforms other state-of-the-art methods.

Abstract With the rapid development of information technology, advanced persistent threats (APTs) have led to numerous serious data breaches and information system disruptions, causing immense losses to governments, businesses, and individuals. APT attack activities are usually carried out stealthily and often require analyzing large amounts of audited data, making it difficult to handle APT attacks promptly. Existing work attempts to improve the handling and detection efficiency of APT attacks based on limited audit data. However, these methods only increase the number of attack samples by finding suspicious entities through rules, ignoring the attack features contained in potential relations between entities. In this paper, we propose a potential relation prediction-based method (APMP) for APT attack detection in few-shot scenarios, which exploits potential relations to find ignored attack features. Specifically, APMP extracts the information between entities and relations in the attack sequence to train the prediction model. The prediction model can predict potential relations between entities and map them into the provenance graph. In this way, APMP complements the potential relations between entities in the provenance graph and captures the attack-related information between entities, improving the results of attack detection. We evaluate APMP using ten real-world public APT attack datasets. The average evaluation precision of APMP attack detection is 100%, with a recall rate of 93.18% and an F1-score of 96.30%. The results show that our proposal can effectively detect APT attacks in few-shot scenarios.

Decentralized storage auditing approaches are designed to ensure data security in dishonest decentralized storage providers. However, the need for data updates introduces new challenges to the design of decentralized storage auditing approaches. Existing approaches can support dynamic auditing for updated files. Unfortunately, they can only deal with block-level updating, which is counter-intuitive and requires conversion from semantic changes to binary changes. Furthermore, existing dynamic auditing approaches require the recalculation of auxiliary auditing information (e.g., auditing authenticators) in data owners, which imposes unnecessary additional burdens on data owners, particularly those with constrained resources in decentralized storage environments. In this paper, we focus on image files and propose iAudit, an efficient pixel-level dynamic image auditing approach in decentralized storage. We first design a novel image authenticator with image pixels for efficient dynamic auditing, which combines convolution operations and polynomial commitment in authenticator construction. Additionally, we build an owner-free dynamic mechanism in dynamic decentralized storage auditing approach by utilizing zero-knowledge proof techniques. In this way, the dynamic operation overheads incurred by auditing can be completely eliminated from the data owners. A prototype of iAudit is implemented, and extensive experimental results demonstrate that iAudit outperforms state-of-the-art works, achieving over a 210× speedup for data owner in dynamic update phase.

Previous model inversion attacks in collaborative inference systems have only been demonstrated on simple models. They failed to produce convincing results on large models or large images. To improve the quality of the model inversion attack, two adaptive generative model inversion attacks are proposed. First, the distributional prior of the publicly available pre-trained Generative Adversarial Nets are used to guide the reconstruction process. Second, previous works only search the latent space of the pre-trained GAN. However, the pre-trained generative model and the target model are trained on different datasets. The distribution shifts of the datasets can lead to inevitable reconstruction errors. To handle the distribution shifts, the pre-trained generative model will be fine-tuned for target instances in the proposed adaptive generative model inversion attack. As a result, the proposed attacks can better handle the distribution shifts. And there is no need to train a separate image prior for each target model. The generators can be used to attack different models trained on different datasets within the same domain. Our extensive experiments demonstrate that the proposed attack achieves up to a 25.7% improvement in PSNR and a 26.4% improvement in SSIM compared to previous methods.

Federated fine-tuning efficiently adapts large pre-trained models to new tasks by using additional data while minimizing re-training costs. This approach enhances data privacy and reduces computational demands but relies on a central server, often cloud-based, which is vulnerable to adversarial attacks that can compromise the aggregation process. We propose${\sf FLAGuard}$, a novel and efficient verification scheme that specifically addresses these challenges in federated Low-Rank Adaptation (LoRA) settings.${\sf FLAGuard}$is the first to introduce a two-stage verification process specifically designed for LoRA-based aggregation. In the first stage, the scheme independently verifies the correctness of the aggregated$A$and$B$matrices. In the second stage, it verifies the multiplication result of the aggregated$A$and$B$matrices, ensuring the correctness of the final LoRA parameters. Additionally, we introduce the Iterative Gradient Sampling and Convolutional Compression (IGSCC) technique, which combines probabilistic sampling with convolutional operations to efficiently reduce the dimensionality of gradient matrices. This enables secure verification without sacrificing model performance. Our comprehensive security analysis of${\sf FLAGuard}$further establishes its reliability in federated learning environments. Extensive experimental results demonstrate that${\sf FLAGuard}$achieves over a$100\times$speedup in the aggregation verification phase and reduces communication overhead by more than 50% compared to state-of-the-art methods.

The general-purpose Segment Anything Model (SAM) is limited by the inherent constraints of RGB sensors, which render it inadequate for challenging real-world scenarios such as adverse lighting conditions and rapid motion. In contrast, event cameras, a novel type of bio-inspired visual sensor, offer distinct imaging advantages, including high temporal resolution and a high dynamic range. The event streams generated by these cameras provide spatiotemporal dynamic cues that are often absent in conventional image frames. To overcome the limitations of RGB-based models, we propose SAM with Event-based Assistance (EvSAM) , a novel RGB-event multi-modal semantic segmentation framework. EvSAM leverages the strong generalization capabilities of SAM while incorporating the complementary characteristics of event data to enhance scene comprehension, particularly under adverse conditions. To address the challenges of fusing two modals (image and event) with large data format discrepancy, we introduce two core components: the Multi-spatiotemporal-scale Patch Alignment Block (MS 2 PAB) and the Event-based Feature Injector (EFInj) for SAM. Specifically, the MS \({}^{2}\) PAB captures spatiotemporal semantic coherence from the event stream and transforms it into a frame-based complementary representation using a multi-spatiotemporal alignment strategy. The EFInj introduces a dynamic event feature update mechanism, wherein the fused features at a given layer guide the adaptive generation of deeper event representations. This process facilitates the integration of RGB spatial semantics with event-based motion cues. Owing to these core designs, EvSAM demonstrates superior performance on event-based semantic segmentation datasets, thereby fully validating its distinct advantages in handling extreme visual scenarios. Furthermore, we extend our model to the task of depth estimation, which further demonstrates its strong generalization ability and scalability for various downstream applications.

When dealing with multi-view data, the heterogeneity of data attributes across different views often leads to label ambiguity. To effectively address this challenge, this paper designs a Multi-View Partial-Label Learning (MVPLL) framework, where each training instance is described by multiple view features and associated with a set of candidate labels, among which only one is correct. The key to deal with such problem lies in how to effectively fuse multi-view information and accurately disambiguate these ambiguous labels. In this paper, we propose a novel approach named CFDM, which explores the consistency and complementarity of multi-view data by multi-view contrastive fusion and reduces label ambiguity by multi-class contrastive prototype disambiguation. Specifically, we first extract view-specific representations using multiple view-specific autoencoders, and then integrate multi-view information through both inter-view and intra-view contrastive fusion to enhance the distinctiveness of these representations. Afterwards, we utilize these distinctive representations to establish and update prototype vectors for each class within each view. Based on these, we apply contrastive prototype disambiguation to learn global class prototypes and accordingly reduce label ambiguity. In our model, multi-view contrastive fusion and multi-class contrastive prototype disambiguation are conducted mutually to enhance each other within a coherent framework, leading to a more ideal classification performance. Experimental results on multiple datasets have demonstrated that our proposed method is superior to other state-of-the-art methods.

Event cameras have gained attention in segmentation due to their higher temporal resolution and dynamic range compared to traditional cameras. However, they struggle with issues like lack of color perception and triggering only at motion edges, making it hard to distinguish objects with similar contours or segment spatially continuous objects. Our work aims to address these often overlooked issues. Based on the assumption that various objects exhibit different motion patterns, we believe that embedding the historical motion states of objects into segmented scenes can effectively address these challenges. Inspired by this, we propose the ESS framework ``Know Where You Are From" (KWYAF), which incorporates past motion cues through spatio-temporal propagation embedding. This framework features two core components: the Sequential Motion Encoding Module (SME) and the Event-Based Reliable Region Selection Mechanism (ER²SM). SMEs construct prior motion features through spatio-temporal correlation modeling for boosting final segmentation, while ER²SM adapts to identify high-confidence regions, embedding motion more precisely through local window masks and reliable region selection. A large number of experiments have demonstrated the effectiveness of our proposed framework in terms of both quantity and quality.

Federated Multi-View Clustering (FMVC) aims to learn a global clustering model from heterogeneous data distributed across different devices, where each device only stores one view of all clustering samples. The key to deal with such problem lies in how to effectively fuse these heterogeneous samples while strictly preserve the data privacy across multiple devices. In this paper, we propose a novel structural graph learning framework named MGCD, which leverages both consistency and diversity of multi-view graph structure across global view-fusion server and local view-specific clients to achieve desired clustering while better preserves data privacy. Specifically, in each local client, we design a dual autoencoder to extract the latent consensuses and specificities of each view, where self-representation construction is introduced to generate the corresponding view-specific diversity graph. In the global server, the consistency implied in uploaded diversity graphs are further distilled and then incorporated into the consistency graph for subsequent cross-view contrastive fusion. During the training process, the server generates a global consistency graph and distributes it to each client for assisting in diversity graph construction, while the clients extract view-specific information and upload it to the server for more reliable consistency graph generation. The ``server-client'' interaction is conducted in an iterative manner, where the consistency implied in each local client is gradually aggregated into the global consistency graph, and the final clustering results are obtained by spectral clustering on the desired global consistency graph. Extensive experiments on various datasets have demonstrated the effectiveness of our proposed method on clustering federated multi-view data.

Salient object detection (SOD) methods for 2D images have great significance in the field of human-computer interaction (HCI). However, as a common data format in HCI, the SOD research in the form of 3D point cloud data remains limited. Previous works commonly treat this task as point cloud segmentation, which perceives all points in the scene for prediction. However, these methods neglect that SOD is designed to simulate human visual perception where human can only see the surfaces rather than occluded point clouds. Thereby, these methods may fail when meet such situations. This paper aims to solve this problem by approximately simulating the perception paradigm of humans towards 3D scenes. Thus, we propose a framework based on the 3D visual point cloud backbone and its multi-view projection named MSV-PCT. Specifically, instead of relying solely on general point cloud learning frameworks, we additionally introduce multi-sparse-view learning branches to supplement the SOD perception. Furthermore, we propose a novel point cloud edge detection loss function to effectively address artifacts, enabling the accurate segmentation of the edges of salient objects from the background. Finally, to evaluate the generalization of point cloud SOD methods, we introduce a new approach to generate simulated PC-SOD datasets from RGBD-SOD data. Experiments on the simulated datasets show that MSV-PCT achieves better accuracy and robustness.

In multi-view multi-label classification (MVML), each object is described by several heterogeneous views while annotated with multiple related labels. The key to learn from such complicate data lies in how to fuse cross-view features and explore multi-label correlations, while accordingly obtain correct assignments between each object and its corresponding labels. In this paper, we proposed an advanced MVML method named VAMS, which treats each object as a bag of views and reformulates the task of MVML as a “view-label” matching selection problem. Specifically, we first construct an object graph and a label graph respectively. In the object graph, nodes represent the multi-view representation of an object, and each view node is connected to its K-nearest neighbor within its own view. In the label graph, nodes represent the semantic representation of a label. Then, we connect each view node with all labels to generate the unified “view-label” matching graph. Afterwards, a graph network block is introduced to aggregate and update all nodes and edges on the matching graph, and further generating a structural representation that fuses multi-view heterogeneity and multi-label correlations for each view and label. Finally, we derive a prediction score for each view-label matching and select the optimal matching via optimizing a weighted cross-entropy loss. Extensive results on various datasets have verified that our proposed VAMS can achieve superior or comparable performance against state-of-the-art methods.

Event-based semantic segmentation (ESS) has attracted researchers' attention recently, as event cameras can solve problems such as under/over-exposure or motion blur that are difficult for RGB cameras to handle. However, event data are noisy and sparse, resulting in difficulties for the model to locate and extract reliable cues from their sparse representations, especially when performing pixel-level tasks. In this paper, we propose a novel framework ESEG to alleviate the dilemma. Given that event signals relate closely to moving edges, instead of proposing complex structures to expect them to recognize those reliable edge regions behind event signals on their own, we introduce the explicit edge-semantic supervision as a reference to let the ESS model globally optimize semantics, considering the high confidence of event data in edge regions. In addition, we propose a fusion module named Density-Aware Dynamic-Window Cross Attention Fusion (D\textsuperscript{2}CAF), in which the density perception, cross-attention, and dynamic window masking mechanisms are jointly imposed to optimize edge-dense feature fusion, leveraging the characteristics of event cameras. Experimental results on DSEC and DDD17 datasets demonstrate the efficacy of the ESEG framework and its core designs.

Reinforcement learning-based knowledge graph reasoning requires complex logical reasoning based on given query relations. Existing methods rely exclusively on delayed reward signals to train the model to perceive the logical reasonableness of the whole reasoning path. This paradigm cannot recognize the logical reasonableness of intermediate reasoning actions, thereby limiting its performance. The logical reasonableness of the same reasoning action changes under different reasoning histories, which makes it difficult for the model to perceive the logical reasonableness of reasoning actions. This paper proposes a Dynamic Logic Alignment-based knowledge graph reasoning method (DLA). DLA dynamically combines the reasoning history and actions, and aligns its logical meaning with the query relation to assess the reasonableness of actions. Firstly, considering that actions have different logical meanings under different history paths, we design a reasoning history-aware dynamic action enhancement mechanism. The mechanism enhances the representation of the current action by injecting the logical composition of history and action at each time step. Secondly, considering that logical composition can provide the reasoning basis for the reinforcement learning agent, we design a query enhancement mechanism for logical composition alignment. The mechanism selects the action with high reasonableness by aligning the logical composition and query relation. Our method has been evaluated on five datasets of different scales, and the experimental results reveal that our method outperforms existing methods.

Multi-view classification aims at unifying the data from multiple views to complementarily enhance the classification performance. Unfortunately, two major problems in multi-view data are damaging model performance. The first is feature heterogeneity, which makes it hard to fuse features from different views. Considering this, we introduce a multi-scale alignment module, including an instance-scale alignment module and a prototype-scale alignment module to mine the commonality from an inter-view perspective and an inter-class perspective respectively, jointly alleviating feature heterogeneity. The second is information redundancy which easily incurs ambiguous data to blur class boundaries and impair model generalization. Therefore, we propose a novel expanded boundary by extending the original class boundary with fuzzy set theory, which adaptively adjusts the boundary to fit ambiguous data. By integrating the expanded boundary into the prototype-scale alignment module, our model further tightens the produced representations and reduces boundary ambiguity. Additionally, compared with the original class boundary, the expanded boundary preserves more margins for classifying unseen data, which guarantees the model generalization. Extensive experiment results across various real-world datasets demonstrate the superiority of the proposed model against existing state-of-the-art methods.

Graph contrastive learning (GCL) aims at narrowing positives while dispersing negatives, often causing a minority of samples with great similarities to gather as a small group. It results in two latent shortcomings in GCL: 1) local cohesion that a class cluster contains numerous independent small groups, and 2) global sparseness that these small groups (or isolated samples) dispersedly distribute among all clusters. These shortcomings make the learned distribution only focus on local similarities among partial samples, which hinders the ability to capture the ideal global structural properties among real clusters, especially high intra-cluster compactness and inter-cluster separateness. Considering this, we design a novel fuzzy boundary by extending the original cluster boundary with fuzzy set theory, which involves fuzzy boundary construction and fuzzy boundary contraction to address these shortcomings. The fuzzy boundary construction dilates the original boundaries to bridge the local groups, and the fuzzy boundary contraction forces the dispersed samples or groups within the fuzzy boundary to gather tightly, jointly mitigating local cohesion and global sparseness while forming the ideal global structural distribution. Extensive experiments demonstrate that a graph auto-encoder with the fuzzy boundary significantly outperforms current state-of-the-art GCL models in both downstream tasks and quantitative analysis.

In multi-view multi-label classification (MVML), each object has multiple heterogeneous views and is annotated with multiple labels. The key to deal with such problem lies in how to capture cross-view consistent correlations while excavate multi-label semantic relationships. Existing MVML methods usually employ two independent components to address them separately, and ignores their potential interaction relationships. To address this issue, we propose a novel Tensorized MVML method named TMvML, which formulates an MVML tensor classifier to excavate comprehensive cross-view feature correlations while characterize complete multi-label semantic relationships. Specifically, we first reconstruct the MVML mapping matrices as an MVML tensor classifier. Then, we rotate the tensor classifier and introduce a low-rank tensor constraint to ensure view-level feature consistency and label-level semantic co-occurrence simultaneously. To better characterize the low-rank tensor structure, we design a new Laplace Tensor Rank (LTR), which serves as a tighter surrogate of tensor rank to capture high-order fiber correlations within the tensor space. By conducting the above operations, our method can easily address the two key challenges in MVML via a concise LTR tensor classifier and achieve the extraction of both cross-view consistent correlations and multi-label semantic relationships simultaneously. Extensive experiments demonstrate that TMvML significantly outperforms state-of-the-art methods.

Hypergraph contrastive learning enables effective representation learning for hypergraphs without requiring labels. However, existing methods typically rely on randomly deleting or replacing nodes during hypergraph augmentation, which may lead to the absence of critical nodes and further disrupt the higher-order structural relationships within augmented hypergraphs. To address this issue, we propose a Critical Node-aware hypergraph contrastive learning method, which is the first attempt to leverage hyperedge prediction to retain critical nodes and accordingly maintain the reliable higher-order structural relationships within augmented hypergraphs. Specifically, we first employ contrastive learning to align the augmented hypergraphs, and then generate hyperedge embeddings to characterize node representations and their structural correlations. During the hyperedge embedding encoding process, we introduce a hyperedge prediction discriminator to score these embeddings, which quantifies the nodes' contributions to identify the critical nodes and maintain the higher-order structural relationships within augmented hypergraphs. Compared with previous studies, our proposed method can effectively alleviate the erroneous deletion or replacement of critical nodes and steadily maintain the inherent structural relationships between original hypergraph and augmented hypergraphs, naturally guiding better hypergraph representations for downstream tasks. Extensive experiments on various tasks demonstrate that our method is significantly superior to state-of-the-art methods.

In recent years, the focus on data privacy and security has intensified, with Secure Multi-party Computation (MPC) providing privacy protection for data and models at the cost of increased computational demands. While existing studies emphasize the computational requirements of non-linear inference, our findings reveal that linear computations can also significantly impact model speed, especially in resourceconstrained environments. In this work, we introduce MPCFLC, an optimization framework for secure inference models. Our innovative two-stage distillation process, which integrates matrix decomposition with non-linear substitution, achieves a$2.52 \times$speedup in inference with negligible performance degradation. Furthermore, our specially crafted distillation method enhances distillation speed by$1.3 \times$, further minimizing accuracy loss. Experiments conducted on the GLUE dataset validate the effectiveness of our proposed approach.

Graph contrastive learning (GCL) aims to learn self-supervised representations by distinguishing positive and negative sample pairs generated from multiple augmented graph views. Despite showing promising performance, GCL still suffers from two critical biases: (1) ***Similarity estimation bias*** arises when feature elements that support positive pair alignment are suppressed by conflicting components within the representation, causing truly positive pairs to appear less similar. (2) ***Semantic shift bias*** occurs when random augmentations alter the underlying semantics of samples, leading to incorrect positive or negative assignments and injecting noise into training. To address these issues, we propose CaliGCL, a GCL model for calibrating the biases by integrating an exponential partitioned similarity measure and a semantics-consistency discriminator. The exponential partitioned similarity computes the similarities among fine-grained partitions obtained through splitting representation vectors and uses exponential scaling to emphasize aligned (positive) partitions while reducing the influence of misaligned (negative) ones. The discriminator dynamically identifies whether augmented sample pairs maintain semantic consistency, enabling correction of misleading contrastive supervision signals. These components jointly reduce biases in similarity estimation and sample pairing, guiding the encoder to learn more robust and semantically meaningful representations. Extensive experiments on multiple benchmarks show that CaliGCL effectively mitigates both types of biases and achieves state-of-the-art performance.

Event cameras, with their capacity to provide high temporal resolution information between frames, are increasingly utilized for video frame interpolation (VFI) in challenging scenarios characterized by high-speed motion and significant occlusion. However, prevalent issues of blur and distortion within the keyframes and ground truth data used for training and inference in these demanding conditions are frequently overlooked. This oversight impedes the perceptual realism and multi-scene generalization capabilities of existing event-based VFI (E-VFI) methods when generating interpolated frames. Motivated by the observation that semantic-perceptual discrepancies between degraded and pristine images are considerably smaller than their image-level differences, we introduce EPA. This novel E-VFI framework diverges from approaches reliant on direct image-level supervision by constructing multilevel, degradation-insensitive semantic perceptual supervisory signals to enhance the perceptual realism and multi-scene generalization of the model's predictions. Specifically, EPA operates in two phases: it first employs a DINO-based perceptual extractor, a customized style adapter, and a reconstruction generator to derive multi-layered, degradation-insensitive semantic-perceptual features ($\mathcal{S}$). Second, a novel Bidirectional Event-Guided Alignment (BEGA) module utilizes deformable convolutions to align perceptual features from keyframes to ground truth with inter-frame temporal guidance extracted from event signals. By decoupling the learning process from direct image-level supervision, EPA enhances model robustness against degraded keyframes and unreliable ground truth information. Extensive experiments demonstrate that this approach yields interpolated frames more consistent with human perceptual preferences. *The code will be released upon acceptance.*

The Unaligned Multi-view Clustering (UMC) aims to learn a discriminative cluster structure from unaligned multi-view data, where the features of samples are not completely aligned across multiple views. Most existing methods usually prioritize employing various alignment strategies to align sample representations across views and then conduct cross-view fusion on aligned representations for subsequent clustering. However, ***due to the heterogeneity of representations across different views, these alignment strategies often fail to achieve ideal view-alignment results, inevitably leading to unreliable alignment-based fusion.*** To address this issue, we propose an alignment-free consistency fusion framework named AF-UMC, which bypasses the traditional view-alignment operation and directly extracts consistent representations from each view to perform global cross-view consistency fusion. Specifically, we first construct a cross-view consistent basis space by a cross-view reconstruction loss and a designed Structural Clarity Regularization (SCR), where autoencoders extract consistent representations from each view through projecting view-specific data to the constructed basis space. Afterwards, these extracted representations are globally pulled together for further cross-view fusion according to a designed Instance Global Contrastive Fusion (IGCF). Compared with previous methods, AF-UMC directly extracts consistent representations from each view for global fusion instead of alignment for fusion, which significantly mitigates the degraded fusion performance caused by undesired view-alignment results while greatly reducing algorithm complexity and enhancing its efficiency. Extensive experiments on various datasets demonstrate that our AF-UMC exhibits superior performance against other state-of-the-art methods.

Online reviews allow consumers to provide detailed feedback on various aspects of items. Existing methods utilize these aspects to model users' fine-grained preferences for specific item features through graph neural networks. We argue that the performance of items on different aspects is important for making precise recommendations, which has not been taken into account by existing approaches, due to lack of data. In this paper, we propose an aspect performance-aware hypergraph neural network (APH) for the review-based recommendation, which learns the performance of items from the conflicting sentiment polarity of user reviews. Specifically, APH comprehensively models the relationships among users, items, aspects, and sentiment polarity by systematically constructing an aspect hypergraph based on user reviews. In addition, APH aggregates aspects representing users and items by employing an aspect performance-aware hypergraph aggregation method. It aggregates the sentiment polarities from multiple users by jointly considering user preferences and the semantics of their sentiments, determining the weights of sentiment polarities to infer the performance of items on various aspects. Such performances are then used as weights to aggregate neighboring aspects. Experiments on six real-world datasets demonstrate that APH improves MSE, Precision@5, and Recall@5 by an average of 2.30%, 4.89%, and 1.60% over the best baseline. The source code and data are available at https://github.com/dianziliu/APH.

ABSTRACT Federated learning (FL) for medical image segmentation poses critical challenges, including non‐IID data distributions, limited access to labelled annotations, and stringent privacy constraints across institutions. To address these, we propose FUSION (Federated Unified Semi‐Supervised Optimisation Network), a novel dual‐path training framework that integrates both Federated Labelled Data Learning (FLDL) and Federated Unlabelled Data Training (FUDT). Central to FUSION is a two‐stage pseudo‐label refinement strategy designed to ensure robustness under real‐world federated constraints. First, synthetic label denoising is performed using Monte Carlo dropout‐based uncertainty estimation, enabling clients to identify and exclude low‐confidence predictions. Second, prototype‐based correction is applied to further refine pseudo‐labels by aligning them with class‐specific feature centroids, mitigating errors caused by domain shifts and inter‐client variability. These refined labels are used for localised training on unlabelled clients, while a dynamic aggregation scheme modulated by a reliability‐based hyperparameter μ adjusts the influence of labelled versus unlabelled clients during global model updates. This tightly coupled interaction between pseudo‐label quality and federated optimisation ensures stability, accelerates convergence, and enhances generalisation across heterogeneous clients. FUSION is evaluated on three diverse datasets: TCGA‐LGG (brain MRI), Kvasir‐SEG (colonoscopy), and UDIAT (ultrasound) and consistently outperforms state‐of‐the‐art FL models in Dice, IoU, HD95, and ASD metrics. Results confirm the critical role of synthetic label refinement in enhancing segmentation accuracy, boundary precision, and model scalability. FUSION provides a technically grounded, privacy‐preserving, and label‐efficient solution for real‐world multi‐institutional medical image segmentation tasks.

Multi-view multi-label classification is a crucial machine learning paradigm aimed at building robust multi-label predictors by integrating heterogeneous features from various sources while addressing multiple correlated labels. However, in real-world applications, concerns over data confidentiality and security often prevent data exchange or fusion across different sources, leading to the challenging issue of data islands. To tackle this problem, we propose a general federated multi-view multi-label classification method, FMVML, which integrates a novel multi-view multi-label classification technique into a federated learning framework. This approach enables cross-view feature fusion and multi-label semantic classification while preserving the data privacy of each independent source. Within this federated framework, we first extract view-specific information from each individual client to capture unique characteristics and then consolidate consensus information from different views on the global server to represent shared features. Unlike previous methods, our approach enhances cross-view fusion and semantic expression by jointly capturing both feature and semantic aspects of specificity and commonality. The final label predictions are generated by combining the view-specific predictions from individual clients and the consensus predictions from the global server. Extensive experiments across various applications demonstrate that FMVML fully leverages multi-view data in a privacy-preserving manner and consistently outperforms state-of-the-art methods.

The rapid development of Industrial IoT (IIoT) has caused the explosion of industrial data, which opens up promising possibilities for data analysis in IIoT networks. Due to the limitation of computation and storage capacity, IIoT devices choose to outsource the collected data to remote cloud servers. Unfortunately, the cloud storage service is not as reliable as it claims, whilst the loss of physical control over the cloud data makes it a significant challenge in ensuring the integrity of the data. Existing schemes are designed to check the data integrity in the cloud. However, it is still an open problem since IIoT devices have to devote lots of computation resources in existing schemes, which are especially not friendly to resource-constrained IIoT devices. In this paper, we propose an efficient storage verification approach for cloud-assisted industrial IoT platform by adopting a homomorphic hash function combined with polynomial commitment. The proposed approach can efficiently generate verification tags and verify the integrity of data in the industrial cloud platform for IIoT devices. Moreover, the proposed scheme can be extended to support privacy-enhanced verification and dynamic updates. We prove the security of the proposed approach under the random oracle model. Extensive experiments demonstrate the superior performance of our approach for resource-constrained devices in comparison with the state-of-the-art.

Federated Learning (FL) represents a distributed machine learning approach, enabling the joint training of a global model through the aggregation of gradients from participating clients without necessitating the exchange of raw data. Prior research has explored methods for verifying the correctness of aggregation in this context and mitigating the overhead associated with the verification process. Nonetheless, the advent of Large Language Models (LLMs), with their parameters numbering in the billions, presents ongoing challenges in devising efficient verification mechanisms in FL for large models. In this paper, we propose an innovative Efficient Verifiable Federated Learning scheme${\sf LaVFL}$, which focusing on addressing the verification challenges incurred by LLM. Specifically, we propose an efficient layer-by-layer verification approach for LLMs by designing a Convolution Gradient Compression (CGC) method without compromising model accuracy. Additionally, to minimize computational and communication overheads, we propose an efficient verification strategy PGS, namely, a Probabilistic Gradient Sampling strategy, which aims to reduce the gradient dimensions for each round of verification while ensuring a high probability of comprehensive verification. We implement a prototype of${\sf LaVFL}$, and extensive experimental results demonstrate that${\sf LaVFL}$achieves over a$300 \times$speedup in the aggregation verification phase and reduces communication overheads by more than 75%, compared to VeriFL under the same experimental setup.

The rapid development of the Industrial Internet of Things (IIoT) has led to an explosion of industrial data. Due to computing and storage capacity limitations, IIoT devices often outsource the collected data to remote cloud servers. Unfortunately, cloud storage and cloud sharing services are not as reliable as they claim to be. Existing schemes aim to check data integrity in the cloud through cloud auditing. However, they suffer from a number of security and privacy vulnerabilities. The challenge of designing a secure storage auditing framework for industrial IoT comes from two aspects: 1) lack of physical protection of data owner IIoT devices; 2) privacy issues due to auditing of sensitive shared data. Inspired by the aforementioned challenges, we design the secure storage audit framework to support flexible cloud data sharing in IIoT: S2A-P2FS. The first contribution in our work is the Polynomial Prefix Message Authentication Code(P2MAC) design. We design an innovative P2MAC data structure as a label, which can simultaneously achieve efficient data verification in cloud data storage and privacy protection in flexible cloud data sharing for cloud auditing. The second contribution is the design of a unique Physical Unclonable Function(PUF) for IIoT. Harsh industrial conditions hinder the stable operation of PUFs. To protect the trustness of IIoT data owners, we propose a robust PUF-based physical protection mechanism for IIoT devices. The key point is that the required key is not stored in the memory of IIoT but hidden within its physical structure. A security analysis was conducted to demonstrate the robustness of S2A-P2FS against known vulnerabilities. A prototype was implemented in a real-world IIoT scenario. Experimental results indicate that, compared to state-of-the-art schemes, S2A-P2FS achieves over a 3x speedup in computational time and requires only 67.5% of the communication cost.

In the task of multiview multilabel (MVML) classification, each instance is represented by several heterogeneous features and associated with multiple semantic labels. Existing MVML methods mainly focus on leveraging the shared subspace to comprehensively explore multiview consensus information across different views, while it is still an open problem whether such shared subspace representation is effective to characterize all relevant labels when formulating a desired MVML model. In this article, we propose a novel label-driven view-specific fusion MVML method named L-VSM, which bypasses seeking for a shared subspace representation and instead directly encodes the feature representation of each individual view to contribute to the final multilabel classifier induction. Specifically, we first design a label-driven feature graph construction strategy and construct all instances under various feature representations into the corresponding feature graphs. Then, these view-specific feature graphs are integrated into a unified graph by linking the different feature representations within each instance. Afterward, we adopt a graph attention mechanism to aggregate and update all feature nodes on the unified graph to generate structural representations for each instance, where both intraview correlations and interview alignments are jointly encoded to discover the underlying consensuses and complementarities across different views. Moreover, to explore the widespread label correlations in multilabel learning (MLL), the transformer architecture is introduced to construct a dynamic semantic-aware label graph and accordingly generate structural semantic representations for each specific class. Finally, we derive an instance-label affinity score for each instance by averaging the affinity scores of its different feature representations with the multilabel soft margin loss. Extensive experiments on various MVML applications have verified that our proposed L-VSM has achieved superior performance against state-of-the-art methods. The codes are available at https://gengyulyu.github.io/homepage/assets/codes/LVSM.zip.

In the task of multiview multilabel (MVML) classification, each object is described by several heterogeneous view features and annotated with multiple relevant labels. Existing MVML methods usually assume that these heterogeneous features are strictly view-aligned, and they directly conduct cross-view information fusion to train a multilabel prediction model. However, in real-world scenarios, such strict view-aligned requirement can be hardly satisfied due to the recurrent spatiotemporal asynchronism when collecting MVML data, which would cause inaccurate multiview fusion results and degrade the classification performance. To address this issue, we propose a generalized nonaligned MVML (GNAM) classification method, which achieves multiview information fusion while aligning cross-view features and accordingly learns a desired multilabel classifier. Specifically, we first introduce a multiorder matching alignment strategy to achieve cross-view feature alignments, where both first-order feature correspondence and second-order structure correspondence are jointly integrated to guarantee the compactness of the view-alignment results. Afterward, a commonality- and individuality-based multiview fusion structure is formulated on the aligned-view features to excavate the consistencies and complementarities across different views, which leads all relevant multiview semantic labels, especially rare labels, to be characterized more comprehensively. Finally, we embed adaptive global label correlations to multilabel classification model to further enhance its semantic expression integrity and develop an alternative algorithm to optimize the whole model. Extensive experimental results have verified that GNAM is significantly superior to other state-of-the-art methods.

Modern software has evolved from delivering software products to web services and applications, which need to be protected by security operation centers (SOC) against ubiquitous cyber attacks. Numerous security alerts are continuously generated every day, which have to be efficiently and correctly processed to identify potential threats. Many AIOps (artificial intelligence for IT operations) approaches have been proposed to (semi-)automate the inspection of alerts so as to reduce manual effort as much as possible. However, due to the ever-complicating attacks, a significant amount of manual work is still required in practice to ensure correct analysis results. In this paper, we propose a Context-Aware cLustering approach for cLassifying sEcurity alErts (CALLEE), which fully exploits the rich relationships among alerts in order to precisely identify similar alerts, significantly reducing the workload of SOC. Specifically, we first design a core conceptual model to capture connections among security alerts, based on which we establish corresponding heterogeneous information networks. Next, we systematically design a set of meta-paths to profile typical alert scenarios precisely, contributing to obtaining the representation of security alerts. We then cluster security alerts based on their contextual similarities, considering the tradeoff between the number of clusters and the homogeneity of each cluster. Finally, security operators only need to manually inspect a limited number of alerts within each cluster, pragmatically reducing their workload while ensuring the accuracy of alert classification. To evaluate the effectiveness of our approach, we collaborate with our industrial partner and pragmatically apply the approach to a real alert dataset. The results show that our approach can reduce the workload of SOC by 99.76%, outperforming baseline approaches. In addition, we further investigate the integration of our proposal with the real business scenario of our industrial partner. The feedback from practitioners shows that CALLEE is pragmatically applicable and helpful in industrial settings.

Video Frame Interpolation (VFI) has witnessed a surge in popularity due to its abundant downstream applications. Event-based VFI (E-VFI) has recently propelled the ad-vancement of VFI. Thanks to the high temporal resolution benefits, event cameras can bridge the informational void present between successive video frames. Most state-of-the-art E-VFI methodologies follow the conventional VFI paradigm, which pivots on motion estimation between consecutive frames to generate intermediate frames through a process of warping and refinement. However, this reliance engenders a heavy dependency on the quality and consis-tency of keyframes, rendering these methods susceptible to challenges in extreme real-world scenarios, such as missing moving objects and severe occlusion dilemmas. This study proposes a novel E-VFI framework that directly synthesize intermediate frames leveraging event-based reference, obviating the necessity for explicit motion estimation and substantially enhancing the capacity to handle motion occlusion. Given the sparse and inher-ently noisy nature of event data, we prioritize the relia-bility of the event-based reference, leading to the development of an innovative event-aware reconstruction strategy for accurate reference generation. Besides, we implement a bi-directional event-guided alignment from keyframes to the reference using the introduced E-PCD module. Finally, a transformer-based decoder is adopted for prediction re-finement. Comprehensive experimental evaluations on both synthetic and real-world datasets underscore the superiority of our approach and its potential to execute high-quality VFI tasks.

Semantic segmentation, a fundamental visual task ubiquitously employed in sectors ranging from transportation and robotics to healthcare, has always captivated the research community. In the wake of rapid advancements in large model research, the foundation model for semantic segmentation tasks, termed the Segment Anything Model (SAM), has been introduced. This model substantially addresses the dilemma of poor generalizability of previous segmentation models and the disadvantage in requiring to retrain the whole model on variant datasets. Nonetheless, segmentation models developed on SAM remain constrained by the inherent limitations of RGB sensors, particularly in scenarios characterized by complex lighting conditions and high-speed motion. Motivated by these observations, a natural recourse is to adapt SAM to additional visual modalities without compromising its robust generalizability. To achieve this, we introduce a lightweight SAM-Event-Adapter (SE-Adapter) module, which incorporates event camera data into a cross-modal learning architecture based on SAM, with only limited tunable parameters incremental. Capitalizing on the high dynamic range and temporal resolution afforded by event cameras, our proposed multi-modal Event-RGB learning architecture effectively augments the performance of semantic segmentation tasks. In addition, we propose a novel paradigm for representing event data in a patch format compatible with transformer-based models, employing multi-spatiotemporal scale encoding to efficiently extract motion and semantic correlations from event representations. Exhaustive empirical evaluations conducted on the DSEC-Semantic and DDD17 datasets provide validation of the effectiveness and rationality of our proposed approach.

Log-based anomaly detection is crucial for software reliability assurance. System logs are semi-structured data containing constant and variable contents, both of which can provide valuable features for anomaly detection. Due to variables being heterogeneous and discrete, there is a lack of effective approaches that can comprehensively incorporate features of variables into log-based anomaly detection. In this paper, we propose VCRLog, an anomaly detection method that mines the relationships among the heterogeneous and discrete variables and extracts important features contributing to anomaly detection. Firstly, considering parsing methods cannot accurately extract variables from logs, we propose a variable extraction method based on domain knowledge. Secondly, to capture and extract the relationship feature among heterogeneous and discrete variables, we design a conceptual model based on system operation to construct variable attributed graph, which can mine important feature vectors by structural embeddings. Finally, considering constants directly express the meaning of logs, we combine relationship vectors with semantic vectors of constants to achieve transformer-based anomaly detection. Experimental results show that our proposed method can accurately detect anomalies and maintain high accuracy as the training data size decreases, outperforming existing methods. Our source code and experimental data are publicly available at https://github.com/Fridaywjy/VCRLog.

Summary Industrial Wireless Sensor Networks (IWSNs) have gained significant popularity for their ability to improve plant productivity and production efficiency through self‐organization and rapid deployment. However, the challenge of achieving reliable and sustainable data transmission remains due to the large amount of heterogeneous data generated by large‐scale IWSNs. In this paper, we present a systematic approach that addresses this challenge by focusing on data transmission clustering strategies, optimal cluster head selection, and routing design. We propose a novel Jointly Optimized Clustering Protocol (JOCP), which enhances cluster head selection by considering multiple critical factors that impact the IWSN life cycle. JOCP incorporates two key modules: the many‐objective clustering model and the double‐layer selection evolutionary algorithm. Specifically, the many‐objective clustering model considers cluster head selection from different perspectives, including maximum node survival cycle, minimum node distance, minimum network overall energy consumption, and balanced cluster energy consumption, with the aim of extending the network life cycle. Additionally, the double‐layer selection evolutionary algorithm optimizes the many‐objective clustering model to select appropriate cluster heads. Through performance verification, we demonstrate that the JOCP protocol effectively enhances the network life cycle and increases the number of surviving nodes compared to baseline clustering algorithms. Our research provides a comprehensive solution to the challenges associated with reliable and sustainable data transmission in large‐scale IWSNs, highlighting the potential for improved performance in industrial applications.

Knowledge graphs (KGs) model entities or concepts and their relations in a structural manner. The incompleteness has turned out to be the main challenge that hinders the application of KG. Recently, reinforcement learning (RL) has been recognized as an effective method to deal with such a challenge, which models research tasks into a sequence decision problem without labels. Although an increasing number of studies investigate and analyze knowledge graphs using reinforcement learning, there lacks a systematic literature review that comprehensively and quantitatively analyzes the landscape of RL-based KG research (RL-KG for short). As a result, researchers may have encountered difficulties in appropriately adopting RL techniques in KG research, even reinventing the wheels. In this paper, we follow the Systematic Literature Review (SLR) methodology to survey, screen, and investigate papers of RL-KG. Specifically, we identify 109 highly related papers from 1542, and systematically investigate them with regard to the following five aspects: (1) to what extent RL-KG have been investigated; (2) what application domains have been covered; (3) what RL techniques have been mainly considered; (4) whether there is a connection between the influence and reproducibility of these papers; (5) what specialized datasets, evaluation metrics, and publication venues have been applied. Through an in-depth analysis of the review results, we systematically and comprehensively identify some significant phenomena and analyze the reasons and difficulties of these phenomena. Based on such analysis, we tentatively propose promising future research topics to promote the RL-KG.

Entity alignment (EA) aims to find equivalent entities in knowledge graphs (KGs) from multiple data sources and is a crucial step in integrating KGs. Recent studies learn the similarity of entity embeddings by aggregating neighboring entities. However, these methods solely compare neighboring entities and do not incorporate the connected relation between an entity and its neighbors. In this paper, we propose a novel Entity and Relation joint Interaction Learning (ERIL) approach, which effectively captures the interaction between entities and relations, enhancing the precision of alignment across different KGs. Specifically, the ERIL model jointly learns the neighborhood features of entities and the spatial structure of relations to train a shared permutation matrix, capturing comprehensive associative relations within KGs. Moreover, a semi-supervised iterative framework is designed to leverage the positive interactions between entities and relations to identify more aligned entities. Extensive experiments are conducted on five benchmark datasets to demonstrate the effectiveness of ERIL compared with existing state-of-the-art EA methods. On DBP15K, our model ERIL outperforms currently available EA methods by 1.9% on Hits@10.

As many countries have promulgated laws to protect users’ data privacy, how to legally use users’ data has become a hot topic. With the emergence of federated learning (FL) (also known as collaborative learning), multiple participants can create a common, robust, and secure machine learning model while addressing key issues in data sharing, such as privacy, security, accessibility, etc. Unfortunately, existing research shows that FL is not as secure as it claims, gradient leakage and the correctness of aggregation results are still key problems. Recently, some scholars try to address these security problems in FL by cryptography and verification techniques. However, there are some issues in this scheme that remain unsolved. First, some solutions cannot guarantee the correctness of the aggregation results. Second, existing state-of-the-art FL schemes have a costly computational and communication overhead. In this article, we propose SVFLC, a secure and verifiable FL scheme with chain aggregation to solve these problems. We first design a privacy-preserving method that can solve the problem of gradient leakage and defend against collusion attacks by semi-honest users. Then, we create a verifiable method based on a homomorphic hash function, which can ensure the correctness of the weighted aggregation results. Besides, the SVFLC can also track users who encounter calculation errors during the aggregation process. Additionally, the extensive experiment results on real-world data sets demonstrate that the SVFLC is efficient, compared with other solutions.

Industrial Internet platforms (IIP) can provide many intelligent services based on the industrial big data stored in clouds. However, the vulnerability of cloud storage can cause data corruption, demanding verifying its integrity. Unfortunately, existing cloud storage verification approaches cannot be directly applied to IIP, since they pose heavy computational burdens on the edge side. In this work, we propose an efficient and dynamic storage verification scheme Edasvic for cloud storage in the IIP. We adopt the polynomial commitment to build an efficient homomorphic authenticator, and further design an authenticator accumulator, which can be efficiently generated with limited computational overheads. In addition, we integrate the dynamic information into the authenticator accumulator to support data dynamics. The security of Edasvic is analyzed under the random oracle model. We conduct extensive experiments to evaluate the performance of Edasvic and compare it with the state-of-the-art approaches. Experimental results affirm that Edasvic is superior to existing solutions in terms of computational efficiency.

In the era emphasizing the privacy of personal data, verifiable federated learning has garnered significant attention as a machine learning approach to safeguard user privacy while simultaneously validating aggregated result. However, there are some unresolved issues when deploying verifiable federated learning in edge computing. Due to the constraint resources, edge computing demands cost saving measurements in model training such as model pruning. Unfortunately, there is currently no protocol capable of enabling users to verify pruning results. Therefore, in this paper, we introduce PrVFL, a verifiable federated learning framework that supports model pruning verification and heterogeneous edge computing. In this scheme, we innovatively utilize zero-knowledge range proof protocol to achieve pruning result verification. Additionally, we first propose a heterogeneous delayed verification scheme supporting the validation of aggregated result for pruned heterogeneous edge models. Addressing the prevalent scenario of performance-heterogeneous edge clients, our scheme empowers each edge user to autonomously choose the desired pruning ratio for each training round based on their specific performance. By employing a global residual model, we ensure that every parameter has an opportunity for training. The extensive experimental results demonstrate the practical performance of our proposed scheme.

The rapid development of Industrial IoT (IIoT) opens up promising possibilities for data analysis and machine learning in IIoT networks. As a distributed paradigm, federated learning (FL) allows numerous IIoT devices to collaboratively train a global model without collecting their local data together in central servers. Unfortunately, a centralized server used to aggregate local gradients can be compromised and forge the result, which incurs the need for aggregation verification. Several approaches focusing on verifying the correctness of aggregation have been proposed. However, it is still an open problem since devices have to devote more computation resources for verification, which are especially not friendly to resource-constrained IIoT devices. Furthermore, verifying weighted aggregation has not been supported in existing approaches. In this paper, we propose an efficient verifiable federated learning approach for IIoT networks, which verifies the aggregation of gradients and requires lowest burden on IIoT devices by introducing zero-knowledge proof techniques. Moreover, our design supports weighted aggregation verification to validate the aggregation of weighted gradients in the cloud server. By comparing the proposed approach with the state-of-the-art schemes including VerifyNet and VeriFL, we demonstrate the superior performance of our approach for resource-constrained devices, which minimizes the computational overheads of the IIoT devices.

Blockchain technology, known for its decentralized and immutable nature, serves as the foundation for various applications. As a prominent application of blockchain, decentralized storage is powered by blockchain technology and is expected to provide a reliable and cost-effective alternative to traditional centralized storage. A major challenge in blockchain-powered decentralized storage is how to guarantee the quality of storage services in decentralized storage nodes (DSNs). Storage auditing can ensure the integrity and security of the stored data. Unfortunately, it incurs additional computational costs for data owners and extra storage overheads for DSNs, which thereby cannot be directly applied to decentralized storage networks consisting of nodes with various computation and storage capacity. In this article, we overcome these problems and minimize additional burdens in storage auditing. We propose EDCOMA, a computation and storage efficient auditing scheme for blockchain-based decentralized storage, in which a double compression method is designed to compress data authenticators using both data and polynomial commitment. To prevent replay attacks on double compression launched by DSNs, we introduce zero knowledge proof and design a compression arithmetic circuit to guarantee the execution of compression operations in DSNs. We analyze the security of EDCOMA under the random oracle model and conduct extensive experiments to evaluate the performance of EDCOMA. Experimental results affirm that EDCOMA outperforms state-of-the-art approaches in both computational and storage efficiency.

Reviews are valuable resources that have been widely researched and used to improve the quality of recommendation services. Recent methods use multiple full embedding layers to model various levels of individual preferences, increasing the risk of the data sparsity issue. Although it is a potential way to deal with this issue that models homophily among users who have similar behaviors, the existing approaches are implemented in a coarse-grained way. They calculate user similarities by considering the homophily in their global behaviors but ignore their local behaviors under a specific context. In this paper, we propose a two-tier shared embedding model (TSE), which fuses coarse- and fine-grained ways of modeling homophily. It considers global behaviors to model homophily in a coarse-grained way, and the high-level feature in the process of each user-item interaction to model homophily in a fine-grained way. TSE designs a whole-to-part principle-based process to fuse these ways in the review-based recommendation. Experiments on five real-world datasets demonstrate that TSE significantly outperforms state-of-the-art models. It outperforms the best baseline by 20.50% on the root-mean-square error (RMSE) and 23.96% on the mean absolute error (MAE), respectively. The source code is available at https://github.com/dianziliu/TSE.git.

Advanced persistent threat (APT) attack leverages various intelligence-gathering techniques to obtain sensitive and critical information, imposing increasing threats to modern software enterprises. However, due to the persistent presence of APT attacks, it is difficult to effectively analyze a large amount of audit data for detecting such attacks, especially for small and medium-sized enterprises (SMEs). This limitation hinders security operation centers (SOC) from promptly handling APT attacks. In this paper, we propose an attack path-based method (APM) for APT attack detection on few-shot learning. Specifically, APM first identifies candidate malicious entities from the provenance graph, contributing to the completion of the missing attack paths. Secondly, we propose a systematic method to exploit potential attack behaviors in the attack path based on the identified candidate malicious entities. We evaluate APM through five APT attacks in realistic environments. Compared to existing baselines, the precision, recall, and F1-score of APM for attack detection increased by 0.28%, 1.64%, and 1.13%, respectively. The results show that our proposal can outperform baseline approaches and effectively detect APT attacks based on few-shot learning.

Smart meters are deployed in the smart grid to achieve bidirectional communication, the control center can monitor, predict energy consumption data in real time, and adjust energy supply dynamically. Unfortunately, real-time data may divulge users’ private information. To protect the privacy of real-time data, data aggregation schemes have been proposed to assist the control center in adjusting the supply to meet users’ electricity demands without sacrificing data privacy. However, extreme weather events and potential attacks may damage the meters and change the structure of the smart grid dynamically, it is important to improve the reliability of the data aggregation scheme. Even if some schemes have improved reliability, but the scalability is poor, they are not suitable for the dynamically changing smart grid network structure. To meet this end, a resilient data aggregation scheme for the smart grid is proposed, which 1) offers better reliability by defending against malicious attacks, group management techniques are proposed, and meters can update their keys when the smart grid network structure changes; 2) affords higher scalability; and 3) enables the control center to make fine-grained adjustments. A prototype implementation shows that the proposed scheme is efficient enough for smart meters.

The deep integration of informatization and industrialization has resulted in an increasingly close connection between supervisory control and data acquisition (SCADA) systems and the Internet. The boundaries of the SCADA system are monitored by industrial smart sensors, which only have limited security protection and face severe security threats. One major threat is that smart sensors are vulnerable to physical attacks because they are often installed in unsafe areas far from plant protection. Under this attack, the data of sensors can be easily tampered with. Moreover, since sensors are resource-constrained physical devices, complex and expensive encryption algorithms are not applicable. In this paper, we design a lightweight industrial smart sensor data stream integrity verification scheme based on physical unclonable function (PUF) for industrial IoT, which can protect the physical security of sensors and the integrity of data streams to ensure the secure transmission of industrial smart sensor data streams. We utilize PUF, fuzzy extractor and bit selection algorithm to generate stable PUF responses. A malicious attacker cannot extract the key information through physical attack. In addition, we design a lightweight integrity verification algorithm with efficient key updating based on lightweight cryptographic primitives, making it suitable for resource-constrained physical devices. We perform the security analysis to demonstrate the security of the scheme to known security vulnerabilities. We implement the proposed scheme and evaluate the performance of our scheme with extensive experiments. The experimental results show the scheme is efficient and superior to existing schemes in computational and communication efficiency.

Partial label learning (PLL) aims to learn a robust multi-class classifier from the ambiguous data, where each instance is given with several candidate labels, among which only one label is real. Most existing methods usually cope with such problem by utilizing a feature similarity graph to conduct label disambiguation. However, these methods construct the feature graph by only employing original features, while the influences of latent outliers and the contributions of label space are regrettably ignored. To tackle these issues, in this article, we propose aPrior KnOwledge ConsTrainedAdaptiveGraph FramEwork (POTAGE) for partial label learning, which utilizes an adaptive graph fused with label information to accurately describe the instance relationship and guide the desired model training. Compared with the feature-induced fixed graph, the adaptive graph is deemed to be more robust and accurate to reveal the intrinsic manifold structure within the data, and the embedding label information is expected to effectively alleviate the label ambiguities and enlarge the gap of label confidences between two instances from different classes. Extensive experiments demonstrate that POTAGE achieves state-of-the-art performance.

-norm equals one. In addition, its neutral properties make it significantly different from the commonly studied vector variables (e.g., the Gaussian vector variables). Due to the aforementioned properties, the conventionally applied linear transformation approaches [e.g., principal component analysis (PCA) and independent component analysis (ICA)] are not suitable for neutral vector variables, as PCA cannot transform a neutral vector variable, which is highly negatively correlated, into a set of mutually independent scalar variables and ICA cannot preserve the bounded property after transformation. In recent work, we proposed an efficient nonlinear transformation approach, i.e., the parallel nonlinear transformation (PNT), for decorrelating neutral vector variables. In this article, we extensively compare PNT with PCA and ICA through both theoretical analysis and experimental evaluations. The results of our investigations demonstrate the superiority of PNT for decorrelating the neutral vector variables.

Recently, efforts have been made to explore introducing music content into deep learning-based music recommendation systems.In previous research, with reference to tasks such as speech recognition, music content is often fed into recommendation models as low-level audio features, such as the Mel-frequency cepstral coefficients.However, unlike tasks such as speech recognition, the audio of music often contains multiple sound sources.Hence, low-level time-domain-based or frequencydomain-based audio features may not represent the music content properly, limiting the recommendation algorithm's performance.To address this problem, we propose a music recommendation model based on chord progressions and attention mechanisms.In this model, music content is represented as chord progressions rather than low-level audio features.The model integrates user song interactions and chord sequences of music and uses an attention mechanism to differentiate the importance of different parts of the song.In this model, to make better use of the historical behavioral information of users, we refer to the design of the neural collaborative filtering algorithm to obtain embedding of users and songs.Under this basis, we designed a chord attention layer to mine users' fine-grained preferences for different parts of the music content.We conducted experiments with a subset of the last.fm-1bdataset.The experimental results demonstrate the effectiveness of the method proposed in this paper.

As network techniques rapidly evolve, attacks are becoming increasingly sophisticated and threatening. Network intrusion detection has been widely accepted as an effective method to deal with network threats. Many approaches have been proposed, exploring different techniques and targeting different types of traffic. Anomaly-based network intrusion detection is an important research and development direction of intrusion detection. Despite the extensive investigation of anomaly-based network intrusion detection techniques, there lacks a systematic literature review of recent techniques and datasets. We follow the methodology of systematic literature review to survey and study 119 top-cited papers on anomaly-based intrusion detection. Our study rigorously and comprehensively investigates the technical landscape of the field in order to facilitate subsequent research within this field. Specifically, our investigation is conducted from the following perspectives: application domains, data preprocessing and attack-detection techniques, evaluation metrics, coauthor relationships, and datasets. Based on the research results, we identify unsolved research challenges and unstudied research topics from each perspective, respectively. Finally, we present several promising high-impact future research directions.

Music content has recently been identified as useful information to promote the performance of music recommendations. Existing studies usually feed low-level audio features, such as the Mel-frequency cepstral coefficients, into deep learning models for music recommendations. However, such features cannot well characterize music audios, which often contain multiple sound sources. In this paper, we propose to model and fuse chord, melody, and rhythm features to meaningfully characterize the music so as to improve the music recommendation. Specially, we use two user-based attention mechanisms to differentiate the importance of different parts of audio features and chord features. In addition, a Long Short-Term Memory layer is used to capture the sequence characteristics. Those features are fused by a multilayer perceptron and then used to make recommendations. We conducted experiments with a subset of the last.fm-1b dataset. The experimental results show that our proposal outperforms the best baseline by [Formula: see text] on HR@10.

The rapid growth of various vehicular apps such as automotive navigation and in-car entertainment has brought the explosion of vehicular data. Such a growth has given rise to a huge challenge of maintaining the quality of cloud storage services for the whole period of storage in vehicular networks. As a result, poor quality of services easily causes data corruption problems and thereby threats vehicular data integrity. Blockchain, a tamper-proofing technique, is considered a promising approach for mitigating data integrity risks in cloud storage. However, existing blockchain-based schemes for auditing long-term cloud data integrity suffer from poor communication performance in a vehicular network. In this study, a blockchain-based offline auditing scheme for cloud storage in the vehicular network is proposed to improve auditing performance. Inspired by the data structure of blockchain, we design an evidence chain to achieve offline auditing, which allows the cloud to spontaneously generate data integrity evidence without communicating with auditors during the evidence generation phase. Furthermore, we extend our scheme to support public and automatic validation based on the smart contract. We prove the security of the proposed scheme under the random oracle model and further provide the performance evaluation by comparing with the state-of-the-art approaches.

With the fast growth of system complexity, it is increasingly difficult to comprehensively analyze security of such large-scale systems, which is a knowledge-intensive task. Although there are various available security knowledge sources, they are not well-connected with each other due to their heterogeneity and unstructured descriptions. In this paper, we propose a systematic approach to construct a comprehensive and reusable knowledge graph in the field of information security. Specifically, we first investigate heterogeneous security knowledge sources and establish a detailed ontology of information security, integrating various security conceptual models. Then, we train a security entity identifier based on active learning to extract security knowledge from unstructured descriptions. Such extracted knowledge is then fused to establish a comprehensive and reusable security knowledge graph based on the unified ontology. Finally, we illustrate the utility of our established knowledge graph with a set of exemplary queries and reasoning rules in the context of a real security scenario.

With the increase of data centers, the number of disks also grows rapidly. Therefore, the prediction of disk failures has become an important task for both academia and industry. Existing prediction schemes predict disk failure in the short prediction horizon or with a short time window. However, these schemes cannot achieve ideal performance for a long prediction horizon with a long time window. In this paper, we proposed a deep learning method that can effectively solve the above problems. We refine the Self-Monitoring, Analysis and Reporting Technology (SMART) attributes by using information entropy to select the most related attributes for prediction. Moreover, we proposed the Multiple Channel Convolutional Neural Network based LSTM (MCCNN-LSTM) model to predict whether disk failures will occur in a given disk in next few days. We further evaluate the MCCNN-LSTM model by comparing it with the state-of-the-art works. Extensive experiments show that our model can improve FDR (Fault Detection Rate) to 99.8% and reduce FAR (False Alarm Rate) to 0.2%.

In the smart grid, meters report their real-time electricity consumption data to a utility supplier, and the utility supplier can adjust its supply accordingly. However, adversaries can infer users' privacy behaviors based on publicly transferred real-time electricity consumption data. Data aggregation schemes protect users' privacy from being leaked. We find two major problems are unsolved: 1) meter failure problem and 2) dynamic membership problem. To solve these problems, we designed a dynamic membership group-based multiple-data aggregation scheme. First, a group-based key establishment scheme is proposed, meters are divided into groups, meters in a group build keys to encrypt their data, the meter failure problem is alleviated. If one group has broken meters, the other groups will not be affected. Second, the dynamic join, dynamic leave, and meter replacement techniques are proposed, and the dynamic membership is achieved by allowing meters to update their keys. The simulation results show a meter's computation cost and communication cost are the minima among the related works, which makes the proposed scheme more suitable for the IoT scenario. Besides, we designed a data encoding method and a data retrieve method, we designed two attacks: 1) “bilinear map pairing attack” and 2) “zero attack.”

In recent years, there have been numerous cyber security issues that have caused considerable damage to the society. The development of efficient and reliable Intrusion Detection Systems (IDSs) is an effective countermeasure against the growing cyber threats. In modern high-bandwidth, large-scale network environments, traditional IDSs suffer from a high rate of missed and false alarms. Researchers have introduced machine learning techniques into intrusion detection with good results. However, due to the scarcity of attack data, such methods’ training sets are usually unbalanced, affecting the analysis performance. In this paper, we survey and analyze the design principles and shortcomings of existing oversampling methods. Based on the findings, we take the perspective of imbalance and high dimensionality of datasets in the field of intrusion detection and propose an oversampling technique based on Generative Adversarial Networks (GAN) and feature selection. Specifically, we model the complex high-dimensional distribution of attacks based on Gradient Penalty Wasserstein GAN (WGAN-GP) to generate additional attack samples. We then select a subset of features representing the entire dataset based on analysis of variance, ultimately generating a rebalanced low-dimensional dataset for machine learning training. To evaluate the effectiveness of our proposal, we conducted experiments based on the NSL-KDD, UNSW-NB15, and CICIDS-2017 datasets. The experimental results show that our method can effectively improve the detection performance of machine learning models and outperform the baselines.

Continuously maintaining software requirements traceability links is essential for managing and evolving software systems. Due to development pressure, traceability links are usually missing during the early development phase in practice, and thus many information retrieval-based approaches have been proposed to automatically recover the traceability links. However, such approaches typically calculate textual similarities among software artifacts without considering specific features of different software artifacts, leading to less accurate results. In this paper, we propose a hybrid approach to recover requirements traceability links, which combines machine learning and logical reasoning to explore features of use cases and code. On one hand, our approach engineers features of use cases and code by taking into account their semantics, based on which a classifier is trained by using supervised learning algorithms. On the other hand, we investigate and leverage the structural information of code to incrementally discover traceability links by defining a list of reasoning rules. We have carried out a series of experiments to compare our approach with state-of-the-art methods, the results of which show that our approach significantly outperforms others.

Summary As an essential component of cloud computing, cloud storage provides flexible data storage services for individuals and organizations. By storing multiple replicas of data in servers, cloud storage providers (CSPs) can improve availability and stability of the cloud storage service. To ensure that all data replicas of a cloud user are intact and completely stored in the CSP, many multi‐replica cloud auditing schemes have been proposed. However, such schemes are predominantly based on the public key infrastructure (PKI), which incurs complex certificate management. In addition, existing schemes do not consider support for sector‐level dynamic auditing. In this paper, we propose a fine‐grained dynamic multi‐replica data auditing scheme that has the following features: (1) it uses ID‐based cryptography to eliminate the cost of certificate management, (2) it supports efficient sector‐level dynamic operations on cloud user data, and (3) it optimizes the challenge algorithm to reduce the computational cost of the third party auditor (TPA). We show that the proposed scheme is provably secure based on a random oracle model. The performance analysis and experiments show the efficiency of the proposed scheme.

In this paper, we propose novel strategies for neutral vector variable decorrelation. Two fundamental invertible transformations, namely, serial nonlinear transformation and parallel nonlinear transformation, are proposed to carry out the decorrelation. For a neutral vector variable, which is not multivariate-Gaussian distributed, the conventional principal component analysis cannot yield mutually independent scalar variables. With the two proposed transformations, a highly negatively correlated neutral vector can be transformed to a set of mutually independent scalar variables with the same degrees of freedom. We also evaluate the decorrelation performances for the vectors generated from a single Dirichlet distribution and a mixture of Dirichlet distributions. The mutual independence is verified with the distance correlation measurement. The advantages of the proposed decorrelation strategies are intensively studied and demonstrated with synthesized data and practical application evaluations.

Social media has become a part of our daily life and we use it for many reasons. One of its uses is to get our questions answered. Given a multitude of social media sites, however, one immediate challenge is to pick the most relevant site for a question. This is a challenging problem because (1) questions are usually short, and (2) social media sites evolve. In this work, we propose to utilize topic specialization to find the most relevant social media site for a given question. In particular, semantic knowledge is considered for topic specialization as it can not only make a question more specific, but also dynamically represent the content of social sites, which relates a given question to a social media site. Thus, we propose to rank social media sites based on combined search engine query results. Our algorithm yields compelling results for providing a meaningful and consistent site recommendation. This work helps further understand the innate characteristics of major social media platforms for the design of social Q&A systems.

Nowadays multimedia social networks are fueled by sensational coverage of sex, violence and crime. In this paper, we provide a normalized cross entropy metric to determine whether a headline is a sensational headline or not by the literal consistency between the headline and its corresponding document. Experiments on a Chinese data set show that the traditional relevancy measurements—vector cosine, relative entropy, likelihood and cross entropy—suffer from strong dependence on text length and are unable to effectively identify sensational headline. The experimental results on both Chinese data sets and English data sets show that our metric can cover the positive effects of high-frequency words and overcome the negative effects of the lengths of the title and the document.

Target searching, i.e. fast locating target objects in images or videos, has attracted much attention in computer vision. A comprehensive understanding of factors influencing human visual searching is essential to design target searching algorithms for computer vision systems. In this paper, we propose a combined model to generate scan paths for computer vision to follow to search targets in images. The model explores and integrates three factors influencing human vision searching, top-down target information, spatial context and bottom-up visual saliency, respectively. The effectiveness of the combined model is evaluated by comparing the generated scan paths with human vision fixation sequences to locate targets in the same images. The evaluation strategy is also used to learn the optimal weighting coefficients of the factors through linear search. In the meanwhile, the performances of every single one of the factors and their arbitrary combinations are examined. Through plenty of experiments, we prove that the top-down target information is the most important factor influencing the accuracy of target searching. The effects from the bottom-up visual saliency are limited. Any combinations of the three factors have better performances than each single component factor. The scan paths obtained by the proposed model are optimal, since they are most similar to the human vision fixation sequences.

Rapid development of mobile communications business leads to greater focus on effective mobile DRM (digital right management) for providing improved content protection. To be able to guarantee DRM policies enforcement, the trusted mobile working environment based on a tamper-resistant hardware module is needed. In this paper, firstly, a construction of the trusted mobile computing based on TPM/TPCM is introduced. Thereafter, an example of DRM authentication scheme in user domain integrated with trusted mobile platform is discussed. Based on the new characters provided by trusted computing platform, the authentication scheme can be simplified, which is safe enough to increase the security of latest mobile DRM framework and promote its interoperability and compatibility.