Retrieval-augmented generation (RAG) is widely adopted for knowledge-intensive tasks, but unverified external knowledge can pose risks such as data injection and retrieval pollution, leading to unexpected generation. Existing defenses rely on patch-based fixes, which limit generalization and increase system latency. To address these issues, we propose RAG2RAG, a framework-level security solution specifically designed for RAG. Inspired by human intuition to reason about what can and cannot be said during RAG phase, RAG2RAG augments the main RAG module with a lightweight RAG-based security expert module composed of two components: (1) a Detective that dynamically retrieves supporting evidence, and (2) a Judge that makes final decisions based on retrieved context. The main and expert modules operate in parallel without causing noticeable delays. Experiments across two languages, six domains, and seven types of poisoning attacks demonstrate that RAG2RAG overall achieves higher accuracy and lower attack success rates than seven mainstream baselines. Furthermore, it integrates seamlessly with various RAG architectures, offering efficient protection across diverse threat scenarios.

The significance of Temporal Knowledge Graphs (TKGs) in Artificial Intelligence (AI) lies in their capacity to incorporate time-dimensional information, support complex reasoning and prediction, optimize decision-making processes, enhance the accuracy of recommendation systems, promote multimodal data integration, and strengthen knowledge management and updates. This provides a robust foundation for various AI applications. To effectively learn and apply both static and dynamic temporal patterns for reasoning, a range of embedding methods and large language models (LLMs) have been proposed in the literature. However, these methods often rely on a single underlying embedding space, whose geometric properties severely limit their ability to model intricate temporal patterns, such as hierarchical and ring structures. To address this limitation, this paper proposes embedding TKGs into projective geometric space and leverages LLMs technology to extract crucial temporal node information, thereby constructing the 5EL model. By embedding TKGs into projective geometric space and utilizing Möbius Group transformations, we effectively model various temporal patterns. Subsequently, LLMs technology is employed to process the trained TKGs. We adopt a parameter-efficient fine-tuning strategy to align LLMs with specific task requirements, thereby enhancing the model's ability to recognize structural information of key nodes in historical chains and enriching the representation of central entities. Experimental results on five advanced TKG datasets demonstrate that our proposed 5EL model significantly outperforms existing models.

Incorporating dynamic characteristics into Graph Neural Networks (GNNs) enhances the understanding of dynamic graph evolution, optimizing temporal-spatial representations for real-world dynamic network problems. However, existing discrete-time dynamic graphs (DTDGs) models face two crucial problems that are difficulty in capturing long-term dependencies and the unbalance of performance and computational efficiency. To address these issues, we propose the AGS-DGNN framework for modeling dynamic graphs which includes two components, one is a dynamic GNN model based on an adaptive gradient smoothing mechanism and meta-learning strategy to generate node embeddings and the other is a Transformer model for these node embedding sequences. Specifically, AGS-DGNN first computes the frame-wise loss for the current snapshot and applies an Exponential Moving Average (EMA) mechanism to smooth the loss gradients, thereby generating local EMA gradients. Then, the local EMA gradients from previous snapshots are aggregated to form a global EMA gradient, enabling effective gradient propagation along the temporal dimension under a meta-learning strategy. Finally, the model adaptively integrates the local and global EMA gradients using a dynamic adjustment factor to update the GNN. Additionally, a Transformer with multi-head self-attention mechanism is applied on the embedding sequence of snapshots to enhance the robustness and performance for the framework. Experiments on six public datasets show the advantage of our AGS-DGNN compared with existing baselines, where it has reached the optimum in sixteen out of eighteen performance metrics.

The proliferation of social media exacerbates information fragmentation, posing challenges to understanding public events. We address the problem of event reconstruction with a novel Multi-view Contrast Event Reconstruction (MCER) model. MCER maximizes feature dissimilarity between different views of the same event using contrastive learning, while minimizing mutual information between distinct events. This aggregates fragmented views to reconstruct comprehensive event representations. MCER employs momentum and weight-sharing encoders in a three-tower architecture with supervised contrastive loss for multi-view representation learning. Due to the scarcity of multi-view public datasets, we construct a new Mul-view-data benchmark.Experiments demonstrate MCER’s superior performance on public data and our Mul-view-data, significantly outperforming selfsupervised methods by incorporating supervised contrastive techniques. MCER advances multi-view representation learning to counter information fragmentation and enable robust event understanding.

Researchers have applied knowledge graph embedding (KGE) techniques with advanced neural network techniques, such as capsule networks, for predicting drug-drug interactions (DDIs) and achieved remarkable results. However, most ignore molecular structure and position features between drug pairs. They cannot model the biomedical field's significant relational mapping properties (RMPs,1-N, N-1, N-N) relation. To solve these problems, we innovatively propose CDHse that consists of two crucial modules: 1) Entity embedding module, we obtain position feature obtained by PubMedBERT and Convolutional Neural Network (CNN), obtain molecular structure feature with Graphic Nuaral Network (GNN), obtain entity embedding feature of drug pairs, and then incorporate these features into one synthetic feature. 2) Knowledge graph embedding module, the synthetic feature is Householder projections and then embedded in the complex vector space for training. In this paper, we have selected several advanced models for the DDIs task and performed experiments on three standard BioKG to validate the effectiveness of CDHse.

Alzheimer's Disease (AD) identification through brain network analysis enhances understandings of the intrinsic neural dynamics within the brain, aiding in the clinical treatment and mechanism research of the brain disease. Current research utilizing functional magnetic resonance imaging (fMRI) time series to study brain effective connectivity networks between different brain regions represents an advanced approach. However, these studies are limited by the brain regions of interest (ROIs) predefined by experts due to the complexity in neuroinformatics on the overly macroscopic scale. To address this issue, we propose a novel method based on adaptive fine-grained causal brain network AFGCBN, which automatically segments ROIs into periodic, micro-level time snippets. In concrete terms, this method consists of two primary components: 1) a fine-grained causal variables learning module that splits the entire time series of each ROI from the brain-region level to the time-snippet level vertices, thereby yielding a fine-grained causal brain network for every subject, and 2) a causal graph fusion model based on Graph Isomorphism Network (GIN) is used to further learn the connectivity of the fine-grained brain network, which is achieved through causal interventions on the coarse-grained brain network. The experimental findings demonstrate that our AFGCBN model excels in constructing causal brain networks for AD diagnosis and delivers superior identification accuracy compared to various current techniques.

Autism spectrum disorder (ASD) is a prevalent neurodevelopmental condition.Prompt recognition and treatment are vital for enhancing the life quality of individuals affected by ASD.However, current research either focus on a single atlas or a simple matrix concatenation combination, neglecting the complex and spatial relationship among the brain regions in different atlases.To tackle this weakness, in this paper, we propose a novel multi-atlas time-series feature fusion model with three steps based on spatial overlap proportion of brain regions to obtain an explainable representation of brain networks, which aims to achieve excellent diagnosis of ASD/TC.Specifically, we formally introduce the concept of spatial overlap and give its measurement, spatial overlap proportion.Then, we fuse the brain regions of multi-atlas to obtain an explainable brain networks of each subject.Finally, the GCN classifier is used to perform the final classification.The experimental results on Autism Brain Imaging Data Exchange (ABIDE) demonstrate that our proposed method achieved an accuracy of 0.771.Overall, our method outperforms SOTA methods in ASD/TC classification.

Graph Neural Networks (GNNs) play a pivotal role in graph representation learning, addressing challenges across diverse applications. Despite their significance, data-driven GNNs often overlook biases, raising fairness concerns. Inspired by counterfactuals, we inquire, ’In graph data, how does removing an edge affect model fairness?’ Existing edge fairness measures lack interpretability. Our focus is on exploring how each edge influences model fairness. We introduce the Movement of Edge Weight (MEW) framework, establishing an interpretable estimation of the influence of each edge on model fairness. Specifically, we improve the interpretability of Probabilistic Distribution Disparity (PDD) as a fairness metric when removing edges by extending the chain rule. To achieve effective model debiasing, we propose deleting the top k training edges with the most significant impact on model bias. Experimental results affirm the effectiveness of our method, demonstrating excellent performance.

Attribute Network Embedding (ANE) is one of the most fundamental problems of graph representation learning that was widely applied in numerous fields such as social computing, chemical molecular analysis, transportation prediction. The essential issue of ANE is to integrate the attribute features and structural features of nodes. However, most current methods heavily depended on label information which is very expensive in the real world. We develop a novel attribute network embedding based on intra-inter view contrastive learning, named I2CL-ANE. To address the issue of dependency on label, we introduce Graph Contrastive Learning (GCL) scheme, a self-supervised learning, into Graph Neural Network (GNN) to reduce the cost. To solve the problem of integration, we introduced intra-view contrastive learning module and inter-view contrastive learning module. The former is utilized to independently extract information from the attribute view and structure view, while the latter is employed to extract shared information in attribute view and structure view. Specially, in inter-view contrastive learning module, we introduce a novel consistent constraint to reduce divergence between two views. Extensive experiments on six benchmark datasets demonstrate the superiority of I2CL-ANE over state-of-the-art baseline methods.

Person re-identification (ReID) aims to match the query images with images in the gallery. However, ReID traditionally focuses on outdoor scenes and standing pose, neglecting the complexities of different poses and indoor environments. These neglects correspond to some challenges: postural differences and background noise which interfere with feature learning and matching. To address these issues, we propose a novel multi-pose ReID method, Semantic- and Pose-Guided Feature Fusion (SPGFF), which integrates semantic-guided and pose-guided features. Specially, a semantic-guided module is employed to incorporate global contextual semantic information into the feature representation. This global contextual semantic information refers to the comprehensive understanding of the entire image, including the relationships of different pixels. By incorporating this information, even there are significant variations in pose, the model is enabled to focus on the most pertinent parts of the image. Meanwhile, the pose-guided module uses pose feature to cleanly disentangle bodily semantic components and selectively match corresponding body parts. To the best of our knowledge, this is the first work to introduce the concept of multi-pose ReID, we have provided a benchmark to address the issue of a lack of publicly available datasets. We demonstrate the effectiveness of our approach on both public and proprietary datasets, showcasing its potential to significantly improve person re-identification in previously overlooked scenarios.

Anomaly detection of time series has been widely used in many practical domains, such as network security, industrial inspection, temperature monitoring. However, because the observed values of time series are disturbed by a large amount of noise to be nonlinear and non-stationary, traditional anomaly detection methods are unable to achieve superior effect. To tackle this problem, considering the influence on the reconstruction of different type time series, we propose an empirical mode decomposition (EMD) plus Attention Variational Auto-Encoder (E-AVAE) model including three parts to realize time series anomaly detection based on self-masked reconstruction error. In the first part, EMD is used to decompose the original time series to obtain Intrinsic Mode Functions (IMFs) and residual function. The second part is to obtain the set of reconstruction error using AVAE model which combines self-attention mechanism, VAE and Long Short-Term Memory (LSTM). Based on density peak clustering, a self-masked reconstruction error iterative clustering algorithm is proposed to detect outliers of abnormal time series in the last part. Experiments on four real time series datasets show that the E-AVAE model has outstanding performance in accuracy and model interpretation.

Deep neural networks have been widely used in remote sensing image segmentation. Nowadays, artificial intelligence methods are increasingly applied to remote sensing feature classification. Although convolutional neural networks (CNNs) are widely used for image segmentation tasks, their global feature extraction with increasing image samples is insufficient. Furthermore, transformer is now being focused on computer vision. However, although transformer can capture the global information of remote sensing images, it cannot adequately model the detailed information of image changes. To comprehensively compensate for the defects of CNNs and the transformer network in feature extraction, this study proposes a semantic segmentation network with multifeature fusion (SWINT-RESNet). This network combines the transformer-extracted global and local features and those of CNNs to improve the accuracy of remote sensing image segmentation. The experiments show that the segmentation performance of SWINT-RESNet is superior for both small and medium sample remote sensing image datasets.

Alzheimer’s disease (AD) is a progressive neurodegenerative disease which has a serious impact on patients’ daily lives. Early detection, diagnosis and treatment of AD remain a major challenge in clinical practice. Current methods typically construct brain functionally connectivity networks based on similarities between regions of interest (ROIs) for each individual, neglecting the potential information from identical ROIs. We contend that considering similarities among identical ROIs holds crucial implications for AD classification. To address this issue, we propose a multi-view brain network construction method, which constructs brain network based on the similarity of identical ROIs for AD diagnosis. Firstly, the time series of the same ROIs of all subjects are extracted and recombined into a new set. For the newly composed set the brain connectivity network is obtained based on the similarity between the time series. Next, this constructed network undergoes graph convolutional network (GCN) processing to produce node embeddings. Finally, these embeddings are extracted for all ROIs within each subject, converted into one-dimensional feature vectors, and utilized for disease detection with support vector machine (SVM). To demonstrate the effectiveness of our method, we evaluate the performance of the proposed method in the widely recognized Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, and the results show that our method exhibits higher performance in identifying AD and MCI compared to existing methods.

Prevalence of fake news has seriously disrupted the information ecosystem and undermined social stability and public trust. It stimulated the development of automatic fake news detection method to tackle this dilemma. Most of the methods can be divided into two types, content-based and propagation-based methods. However, these methods overlook the historical background information of the news events contained in the target news in varying degrees. The environment constructed by the historical context of events related to a news event can reflect the direction of the public’s expectations of the current event with respect to future developments. The expectation can be used to study the direction of fake news that has not been fully explored. Therefore, considering the influence of public expectation, we propose a general fake news detection method based on cross-domain coarse-fine grained environments referred to as CFGE including three parts. Specifically, we first construct a cross-domain coarse-fine grained environment with the news related to the domains which were contained in the target news. Then the coarse-grained embedding and the fine-grained embedding of cross-domain environment are extracted respectively by the proposed environment information capture modules. Finally, based on the gate fusion method, coarse-grained embedding and fine-grained embedding are fused to predict fake news. Extensive experimentation substantiates the superiority of CFGE compared to alternative models, further affirming the efficacy of coarse-fine grained environments.

Despite the significant advances in RGB-D scene recognition, there are several major limitations that need further investigation. For example, simply extracting modal-specific features neglects the complex relationships among multiple modalities of features. Moreover, cross-modal features have not been considered in most existing methods. To address these concerns, we propose to integrate the tasks of cross-modal matching and modal-specific recognition, termed as Matching-to-Recognition Network (MRNet). Specifically, the cross-modal matching network enhances the descriptive power of the recognition network via a layer-wise semantic loss. The recognition network obtains multi-modal features from a two-stream CNN: global features are obtained by a higher-layer of a CNN to preserve the semantic content, and local layout features are learned by the graph attention network, thus better capturing the key object regions and modelling their relationships. Extensive experiments results demonstrate the MRNet achieves superior performance to state-of-the-art methods, especially for recognition solely based on single modality.

Recently, action recognition has achieved impressive performance, mainly due to the aid of deep convolutional neural networks and large datasets. Traditionally, most efforts in action recognition have focused on capturing motion information by dense optical flow, but optical flow extraction is very time-consuming. Moreover, prior arts seek to improve accuracy but neglect the part-whole relationship between objects in videos, which may be self-defeating and even deteriorate the performance of methods. To circumvent the above challenges, we present a novel collaborative multipath capsule network (CMCN) for action recognition. In particular, we propose a plug-and-play collaborative multipath block containing spatiotemporal, channel, and motion units, which are complementary and crucial information for action recognition. We exploit the interaction of these three units and selectively emphasize informative spatial-temporal motion to reduce the expensive computational costs. Subsequently, we explore a new capsule voting procedure to reduce the computation used in the capsule dynamic routing mechanism. The critical insight is that the same type of capsules simulates the same entity in different positions, and their voting results should be consistent. This strategy lessens the number of learning parameters that backward pass in the training process, and thus strengthens part-whole relationships in a video. Extensive experiments on multiple real-world datasets for action recognition demonstrate that our model significantly outperforms state-of-the-art models.

Recently, with the development of temporal knowledge graph technology, more and more Temporal Knowledge Graph Embedded (TKGE) models have been developed. The effectiveness of TKGE largely depends on the ability to model intrinsic relation patterns and capture specific information about entities and relations. However, existing approaches can capture only some of them with insufficient modeling capacity, and none has a “deep” architecture for modeling the entries in a quadruple at the same dimension. In this article, we propose a more powerful KGE framework named DuCape , which combines a dual quaternion and capsule network in modeling for the first time to make up for the defects of existing TKGE models. In dual quaternion vector space, the head entity learns a k -dimensional rigid transformation parametrized by relation and time, falling near its corresponding tail entity. Further, we employ the embeddings of entities, relations, and time trained from dual quaternion vector space as the input to capsule networks. Experimental results on several basic datasets show that the DuCape model constructed in this article is superior to existing state-of-the-art models.

Support vector machine (SVM) is a fundamental machine learning method with solid mathematical theory and high effectiveness in many applications. Because distributed datasets are difficult to centralize, SVM is hard to be computed by using traditional algorithms in distributed environment. Meanwhile most of existing distributed SVM methods are suffering in very time-consuming. The dilemma of existing distributed SVM methods has hindered their application in a great deal of domain. In this paper, we focus on the improvement of training efficiency for distributed SVM by proposing a distributed SVM method with distributed loss (namely DL-DSVM). We firstly construct an optimization problem of distributed SVM based on distributed loss. Then, considering constrains in distributed environment, we propose a fast training method to solve the optimization problem based on the local optimal solution. Comprehensive experimental results show that DL-DSVM has an excellent performance in time complexity and robustness, and no significant decline in other aspects.

Privacy leakage is a major hidden danger for the healthy and orderly development of e-commerce. In this paper, we analyse the current situation of the information leakage. The main reason lies in that the customer information is stored and displayed in plaintext during the e-commerce process. We present a customer privacy protection platform on the basis of the analysis, where such technical measures as information segmentation, data encryption and access authorisation are taken. With the customer privacy protection platform, the customer information is not stored and displayed in plaintext any more, which is replaced with two-dimensional code or barcode. The traceability helps to prevent the leakage and abuse of the consumer information as much as possible. Thus, the traditional operation process of e-commerce is then reengineered and the customer information is only used on demand.

A least squares support vector machine (LS-SVM) offers performance comparable to that of SVMs for classification and regression. The main limitation of LS-SVM is that it lacks sparsity compared with SVMs, making LS-SVM unsuitable for handling large-scale data due to computation and memory costs. To obtain sparse LS-SVM, several pruning methods based on an iterative strategy were recently proposed but did not consider the quantity constraint on the number of reserved support vectors, as widely used in real-life applications. In this article, a noniterative algorithm is proposed based on the selection of globally representative points (global-representation-based sparse least squares support vector machine, GRS-LSSVM) to improve the performance of sparse LS-SVM. For the first time, we present a model of sparse LS-SVM with a quantity constraint. In solving the optimal solution of the model, we find that using globally representative points to construct the reserved support vector set produces a better solution than other methods. We design an indicator based on point density and point dispersion to evaluate the global representation of points in feature space. Using the indicator, the top globally representative points are selected in one step from all points to construct the reserved support vector set of sparse LS-SVM. After obtaining the set, the decision hyperplane of sparse LS-SVM is directly computed using an algebraic formula. This algorithm only consumes O(N2) in computational complexity and O(N) in memory cost which makes it suitable for large-scale data sets. The experimental results show that the proposed algorithm has higher sparsity, greater stability, and lower computational complexity than the traditional iterative algorithms.

The semisupervised least squares support vector machine (LS-S3VM) is an important enhancement of least squares support vector machines in semisupervised learning. Given that most data collected from the real world are without labels, semisupervised approaches are more applicable than standard supervised approaches. Although a few training methods for LS-S3VM exist, the problem of deriving the optimal decision hyperplane efficiently and effectually has not been solved. In this paper, a fully weighted model of LS-S3VM is proposed, and a simple integer programming (IP) model is introduced through an equivalent transformation to solve the model. Based on the distances between the unlabeled data and the decision hyperplane, a new indicator is designed to represent the possibility that the label of an unlabeled datum should be reversed in each iteration during training. Using the indicator, we construct an extended candidate set consisting of the indices of unlabeled data with high possibilities, which integrates more information from unlabeled data. Our algorithm is degenerated into a special scenario of the previous algorithm when the extended candidate set is reduced into a set with only one element. Two strategies are utilized to determine the descent directions based on the extended candidate set. Furthermore, we developed a novel method for locating a good starting point based on the properties of the equivalent IP model. Combined with the extended candidate set and the carefully computed starting point, a fast algorithm to solve LS-S3VM quasi-optimally is proposed. The choice of quasi-optimal solutions results in low computational cost and avoidance of overfitting. Experiments show that our algorithm equipped with the two designed strategies is more effective than other algorithms in at least one of the following three aspects: 1) computational complexity; 2) generalization ability; and 3) flexibility. However, our algorithm and other algorithms have similar levels of performance in the remaining aspects.

The last few years have witnessed the emergence and evolution of a vibrant research stream on a large variety of online social media network (SMN) platforms. Recognizing anonymous, yet identical users among multiple SMNs is still an intractable problem. Clearly, cross-platform exploration may help solve many problems in social computing in both theory and applications. Since public profiles can be duplicated and easily impersonated by users with different purposes, most current user identification resolutions, which mainly focus on text mining of users’ public profiles, are fragile. Some studies have attempted to match users based on the location and timing of user content as well as writing style. However, the locations are sparse in the majority of SMNs, and writing style is difficult to discern from the short sentences of leading SMNs such as Sina Microblog and Twitter. Moreover, since online SMNs are quite symmetric, existing user identification schemes based on network structure are not effective. The real-world friend cycle is highly individual and virtually no two users share a congruent friend cycle. Therefore, it is more accurate to use a friendship structure to analyze cross-platform SMNs. Since identical users tend to set up partial similar friendship structures in different SMNs, we proposed the Friend Relationship-Based User Identification (FRUI) algorithm. FRUI calculates a match degree for all candidate User Matched Pairs (UMPs), and only UMPs with top ranks are considered as identical users. We also developed two propositions to improve the efficiency of the algorithm. Results of extensive experiments demonstrate that FRUI performs much better than current network structure-based algorithms.

Mining global patterns on big data distributed in many remote locations is a challenging task since transmitting big data in different remote data servers to the central server is prohibitively expensive. In this paper, we present a succinct distributive big data clustering algorithm based on local-remote coordination (DBDC-LRC) that aims to reduce the cost of big data transmission while maintaining an acceptable overall clustering accuracy. The algorithm is divided into three phases. In the first phase, the idea of Canopy algorithm is improved in the search for representative points with a clustering assumption that the decision boundary should lie in a low-density region, during which controllable thresholds are optimized. Noticing that in data mining a hyperellipsoid is more adaptable in shaping unknown data than a hypercube, we employ Mahalanobis distance as opposed to the Euclidean distance in determining the representative points in different remote data servers. Given that only a limited number of representative points, instead of all the remote data, are transmitted to the central server for clustering, the transmitting cost is reduced significantly. In the second phase, a weighted clustering method is used in mining the global patterns from the gathered information of representative points from various remote data servers. In the third phase, the mined global patterns are sent back to the original remote server and the related data are labeled with the same patterns according their representative points nearby. In this phase, Bayesian method is used to resolve the conflicts that one point is covered by several representative points in its neighborhood. Experiments show that DBDC-LRC is highly suitable for mining patterns from distributive big data, and the advantages of this approach include low cost, high accuracy, high robustness, and good expansibility.