Nan Sun, Xixun Lin*, Zhiheng Zhou, Yanmin Shang, Zhenlin Cheng, Yanan Cao.

Abstract: Recently, Out-of-distribution (OOD) detection in dynamic graphs, which aims to identify whether incoming data deviates from the distribution of the in-distribution (ID) training set, has garnered considerable attention in security-sensitive fields. Current OOD detection paradigms primarily focus on static graphs and confront two critical challenges: i) high bias and high variance caused by single-point estimation, which makes the predictions sensitive to randomness in the data; ii) score homogenization resulting from the lack of OOD training data, where the model only learns ID-specific patterns, resulting in overall low OOD scores and a narrow score gap between ID and OOD data. To tackle these issues, we first investigate OOD detection in dynamic graphs through the lens of Evidential Deep Learning (EDL). Specifically, we propose EviSEC, an innovative and effective OOD detector via Evidential Spectrum-awarE Contrastive Learning. We design an evidential neural network to redefine the output as the posterior Dirichlet distribution, explaining the randomness of inputs through the uncertainty of distribution, which is overlooked by single-point estimation. Moreover, spectrum-aware augmentation module generates OOD approximations to identify patterns with high OOD scores, thereby widening the score gap between ID and OOD data and mitigating score homogenization. Extensive experiments on real-world datasets demonstrate that EviSAC effectively detects OOD samples in dynamic graphs.

Yongxuan Wu, Yang Liu, Xixun Lin, Hong Zhou, Yanan Cao, Lixin Zou, Yanmin Shang, Yanbing Liu.

Abstract: Cross-domain recommendation (CDR) has gained significant attention for its ability to address data sparsity issue. However, most existing CDR methods focus primarily on improving recommendation accuracy while largely overlooking fairness considerations, which can lead to biased outcomes and unfair treatment of different user groups. To solve this critical problem, we investigate whether fairness can be transferred from the source domain to the target domain. Our analysis suggests that fairness can be effectively transferred if the fairness of the source domain is ensured and the distributions of the source and target domains are well aligned. Based on this, we propose the \textbf{FairCDR}, a novel framework that can achieve the knowledge transfer of fairness and user preferences simultaneously. FairCDR owns two phases: single-domain fairness guarantee and inter-domain distribution alignment. In the first phase, we employ an adversarial learning-based recommender (\textbf{ALR}) to disentangle user preferences from sensitive attributes in the source domain. In the second phase, we introduce a new mutual learning-based diffusion model (\textbf{MLDiff}), which engages in mutual learning with ALR to progressively align the distributions of the source and target domains. This improves ALR’s adaptability to distribution shifts, ultimately ensuring fairness and recommendation performance in the target domain. Extensive experiments on multiple real-world cross-domain datasets demonstrate that FairCDR surpasses existing strong baselines in both fairness and recommendation quality.

Xixun Lin, Rui Liu, Yanan Cao, Lixin Zou, Qian Li, Yongxuan Wu, Yang Liu, Dawei Yin, Guandong Xu.

Abstract: Multimodal recommendation, which utilizes rich multimodal information to learn user preferences, has attracted significant attention. Most works focus on designing powerful encoders for extracting multimodal features, and simply aggregate the learned features together to make prediction. Consequently, they have a limited capacity to learn the inter-modality knowledge including the modality-shared and modality-unique knowledge. In fact, learning the modality-shared knowledge enables us to align cross-modality data for fusing heterogeneous modality features. Learning the modality-unique knowledge is equally important when recommendation tasks only involve a small amount of shared features and the necessary information is contained within specific modality. In this paper, we propose Contrastive Modality-Disentangled Learning (CMDL) to overcome this critical limitation. CMDL exactly captures the inter-modality knowledge by achieving modality disentanglement. Specifically, CMDL first disentangles the initial representation into the modality-invariant and modality-specific representations. Afterwards, CMDL introduces a novel manner of contrastive learning to approximate the MI upper bounds for achieving disentanglement regularization. Building upon the proposed regularization, CMDL encourages the modality-invariant and modality-specific representations to capture the modality-shared and modality-unique knowledge respectively and to be statistically independent to each other. Empirically, extensive experiments are conducted on benchmark datasets, demonstrating the superior performance of CMDL compared with strong multimodal recommenders.

Xixun Lin, Yanan Cao, Nan Sun, Lixin Zou, Chuan Zhou, Peng Zhang, Shuai Zhang, Ge Zhang, Jia Wu.

Abstract: Graph-level out-of-distribution (OOD) detection, which attempts to identify OOD graphs originated from an unknown distribution, is a vital building block for safety-critical applications in Web and society. Current approaches concentrate on how to learn better graph representations, but fail to provide any statistically guarantee on detection results, therefore impeding their deployments in the scenario where detection errors would result in serious consequences. To overcome this critical issue, we propose the Conformal Graph-level Out-of-distribution Detection (CGOD), extending the theory of conformal prediction to graph-level OOD detection with a rigorous control over the false positive rate. In CGOD, we develop a new aggregated non-conformity score function based on the proposed adaptive data augmentation. Through the guidance from two designed metrics, i.e., score consistency and representation diversity, our augmentation strategy can generate multiple non-conformity scores, and aggregating these generated non-conformity scores together is robust to the misleading information. Meanwhile, our score function can perceive the subsequent process of conformal inference, enabling the aggregated non-conformity score to be adaptive to different input graphs and deriving a more accurate score estimation. We conduct experiments on multiple real-world datasets with different empirical settings. Extensive results and model analyses demonstrate the superior performance of our approach over several competitive baselines.

Chuancheng Song, Xixun Lin, Hanyang Shen, Yanmin Shang, Yanan Cao.

Abstract: Graph anomaly detection has attracted significant attention due to its critical applications, such as identifying money laundering in financial systems and detecting fake reviews on social networks. However, two major challenges persist: (1) anomaly detection at the node, edge, and graph levels is often addressed in isolation, hindering the integration of complementary information to identify anomalies arising from collective behaviors; and (2) the inherent label sparsity in graph data, coupled with the difficulty of obtaining high-quality annotations, exacerbates bias in detection. To address these challenges, we propose UniFORM, a unified self-supervised anomaly detection framework comprising three modules: UIO, UMC and UPL. UIO unifies node-, edge-, and graph-level tasks from a subgraph perspective, leveraging an energy-based GNN for iterative multi-granular anomaly detection. UMC enhances meta-learning through contrastive learning and employs Langevin dynamics to generate phantom samples as substitutes for anomalous data, reducing reliance on labeled data. UPL design unified loss in intra-inter perspectives. Extensive experiments on real-world datasets demonstrate that UniFORM significantly outperforms state-of-the-art methods across multiple granularities.

Yanan Cao, Xixun Lin*, Yongxuan Wu, Fengzhao Shi, Yanmin Shang, Qingfeng Tan, Chuan Zhou, Peng Zhang.

Abstract: Knowledge Graph Embedding (KGE) aims to learn representations of entities and relations of knowledge graph (KG). Recently Graph Neural Networks (GNNs) have gained great success on KGE, but for the reason behind it, most views simply attribute to the well learning of knowledge graph structure, which still remains a limited understanding of the internal mechanism. In this work, we first study a fundamental problem, i.e., what are the important factors for GNNs to help KGE. To investigate this problem, we discuss the core idea of current GNN models for KG, and propose a new assumption of relational homophily that connected nodes possess similar features after relation’s transforming, to explain why aggregating neighbors with relation can help KGE. Based on the model and empirical analyses, we then introduce a novel data-centric framework for applying GNNs to KGE called KSG-GNN. In KSG-GNN, we construct a new graph structure from KG named Knowledge Similarity Graph (KSG), where each node connects with its similar nodes as neighbors, and then we apply GNNs on this graph to perform KGE. Instead of following the relational homophily assumption in KG, KSG aligns with homogeneous graphs that can directly satisfy homophily assumption. Hence, any GNN developed on homogeneous graphs like GCN, GAT, GraphSAGE, etc., can be applied out-of-the-box as KSG-GNN without modification, which provides a more general and effective GNN paradigm. Finally, we conduct extensive experiments on two benchmark datasets, i.e., FB15k-237 and WN18RR, demonstrating the superior performance of KSG-GNN over multiple strong baselines.

Fengzhao Shi, Yanan Cao, Ren Li, Xixun Lin, Yanmin Shang, Chuan Zhou, Jia Wu, Shirui Pan.

Abstract: Graph Neural Networks (GNNs) have achieved remarkable success in diverse real-world applications. Traditional GNNs are designed based on homophily, which leads to poor performance under heterophily scenarios. Most current solutions deal with heterophily mainly by modeling the heterophily edges as data noises or high-frequency signals, treating all heterophilic edges as being of the same semantic. Consequently, they ignore the rich semantic information of these edges in heterophily graphs. To overcome this critic problem, we propose a novel GNN model based on relation vector translation named as Variational Relation Vector Graph Neural Network (VR-GNN). VR-GNN models relation generation and graph aggregation into an end-to-end model based on a variational inference framework. To be specific, the encoder utilizes the structure, feature and label to generate a fine-grained relation vector for each edge, which aims to infer its implicit semantic information. The decoder incorporates the generated relation vectors into the message-passing framework for deriving better node representations. We conduct extensive experiments on eight real-world datasets with different homophily-heterophily properties to verify model effectiveness. Extensive experimental results show that VR-GNN gains consistent and significant improvements against existing strong GNN methods under heterophily and competitive performance under homophily.

Xixun Lin, Wenxiao Zhang, Fengzhao Shi, Chuan Zhou, Lixin Zou, Xiangyu Zhao, Dawei Yin, Shirui Pan, Yanan Cao

Abstract: Graph neural networks (GNNs) have advanced the state of the art in various domains. Despite their remarkable success, the uncertainty estimation of GNN predictions remains under-explored, which limits their practical applications especially in risk-sensitive areas. Current works suffer from either intractable posteriors or inflexible prior specifications, leading to sub-optimal empirical results. In this paper, we present graph neural stochastic diffusion (GNSD), a novel framework for estimating predictive uncertainty on graphs by establishing theoretical connections between GNNsand stochastic partial differential equation. GNSDrepresents a GNN-based parameterization of the proposed graph stochastic diffusion equation which includes a Q-Wiener process to model the stochastic evolution of node representations. GNSDintroduces a drift network to guarantee accurate prediction and a stochastic forcing network to model the propagation of epistemic uncertainty among nodes. Extensive experiments are conducted on multiple detection tasks, demonstrating that GNSD yields the superior performance over existing strong approaches.

Yu Liu, Yanan Cao, Shi Wang, Qingyue Wang, Guanqun Bi

Abstract: Answering complex logical queries over knowledge graphs (KGs) is a fundamental yet challenging task. Recently, query representation has been a mainstream approach to complex logical reasoning, making the target answer and query closer in the embedding space. However, there are still two limitations. First, prior methods model the query as a fixed vector, but ignore the uncertainty of relations on KGs. In fact, different relations may contain different semantic distributions. Second, traditional representation frameworks fail to capture the joint distribution of queries and answers, which can be learned by generative models that have the potential to produce more coherent answers.To alleviate these limitations, we propose a novel generative model, named DiffCLR, which exploits the diffusion model for complex logical reasoning to approximate query distributions. Specifically, we first devise a query transformation to convert logical queries into input sequences by dynamically constructing contextual subgraphs. Then, we integrate them into the diffusion model to execute a multi-step generative process, and a structure-enhanced self-attention is further designed for incorporating the structural features embodied in KGs. Experimental results on two benchmark datasets show our model effectively outperforms state-of-the-art methods, particularly in multi-hop chain queries with significant improvement.

Yuchen Zhou, Yanan Cao, Yongchao Liu, Yanmin Shang, Peng Zhang, Zheng Lin, Yun Yue, Baokun Wang, Xing Fu, Weiqiang Wang

Abstract: Data augmentation has been widely studied as it can be used to improve the generalizability of graph representation learning models. However, existing works focus only on the data augmentation on homogeneous graphs. Data augmentation for heterogeneous graphs remains under-explored. Considering that heterogeneous graphs contain different types of nodes and links, ignoring the type information and directly applying the data augmentation methods of homogeneous graphs to heterogeneous graphs will lead to suboptimal results. In this paper, we propose a novel Multi-Aspect Heterogeneous Graph Augmentation framework named MAHGA. Specifically, MAHGA consists of two core augmentation strategies: structure-level augmentation and metapath-level augmentation. Structure-level augmentation pays attention to network schema aspect and designs a relation-aware conditional variational auto-encoder that can generate synthetic features of neighbors to augment the nodes and the node types with scarce links. Metapath-level augmentation concentrates on metapath aspect, which constructs metapath reachable graphs for different metapaths and estimates the graphons of them. By sampling and mixing up based on the graphons, MAHGA yields intra-metapath and inter-metapath augmentation. Finally, we conduct extensive experiments on multiple benchmarks to validate the effectiveness of MAHGA. Experimental results demonstrate that our method improves the performances across a set of heterogeneous graph learning models and datasets.

Yuchen Zhou, Yanan Cao, Yanmin Shang, Chuan Zhou, Shirui Pan, Zheng Lin, Qian Li

Abstract: Recommender systems which captures dynamic user interest based on time-ordered user-item interactions plays a critical role in the real-world. Although existing deep learning-based recommendation systems show good performances, these methods have two main drawbacks. Firstly, user interest is the consequence of the coaction of many factors. However, existing methods do not fully explore potential influence factors and ignore the user-item interaction formation process. The coarse-grained modeling patterns cannot accurately reflect complex user interest and leads to suboptimal recommendation results. Furthermore, these methods are implicit and largely operate in a black-box fashion. It is difficult to interpret their modeling processes and recommendation results. Secondly, recommendation datasets usually exhibit scale-free distributions and some existing recommender systems take advantage of hyperbolic space to match the data distribution. But they ignore that the operations in hyperbolic space are more complex than that in Euclidean space which further increases the difficulty of model interpretation. To tackle the above shortcomings, we propose an Explainable Hyperbolic Temporal Point Process for User-Item Interaction Sequence Generation (EHTPP). Specifically, EHTPP regards each user-item interaction as an event in hyperbolic space and employs a temporal point process framework to model the probability of event occurrence. Considering that the complexity of user interest and the interpretability of the model,EHTPP explores four potential influence factors related to user interest and uses them to explicitly guide the probability calculation in the temporal point process. In order to validate the effectiveness of EHTPP, we carry out a comprehensive evaluation of EHTPP on three datasets compared with a few competitive baselines. Experimental results demonstrate the state-of-the-art performances of EHTPP.

Yuchen Zhou, Yanmin Shang, Yanan Cao, Qian Li, Chuan Zhou, Guandong Xu.

Abstract: Attributed graph embedding aims to learn node representation based on the graph topology and node attributes. The current mainstream GNN-based methods learn the representation of the target node by aggregating the attributes of its neighbor nodes. These methods still face two challenges: (1) In the neighborhood aggregation procedure, the attributes of each node would be propagated to its neighborhoods which may cause disturbance to the original attributes of the target node and cause over-smoothing in GNN iteration. (2) Because the representation of the target node is derived from the attributes and topology of its neighbors, the attributes and topological information of each neighbor have different effects on the representation of the target node. However, this different contribution has not been considered by the existing GNN-based methods. In this paper, we propose a novel GNN model named API-GNN (Attribute Preserving Oriented Interactive Graph Neural Network). API-GNN can not only reduce the disturbance of neighborhood aggregation to the original attribute of target node, but also explicitly model the different impacts of attribute and topology on node representation. We conduct experiments on six public real-world datasets to validate API-GNN on node classification and link prediction. Experimental results show that our model outperforms several strong baselines over various graph datasets on multiple graph analysis tasks.

Yuchen Zhou, Yanan Cao, Yanmin Shang, Chuan Zhou, Chuancheng Song, Fengzhao Shi, Qian Li

Abstract: Graph meta-learning which is used to deal with graph few-shot learning attracts more and more research interests. Existing graph meta-learning methods mainly focus on capturing node-level relations, but they ignore task-level relations which are beneficial for improving the performance of few-shot node classification. Furthermore, contrastive learning which can learn knowledge without labeled data is suitable for few-shot scenario, but existing graph few-shot learning methods have never exploited it. To tackle above problems, in this paper, we combine conventional graph meta-learning framework with graph contrastive learning and propose a novel joint model named -T––asklevel -R––elations Modelling for -G––raph M–––eta-learning (TRGM). By constructing auxiliary contrastive pretext tasks, TRGM can fully capture the inter-task relations (task correlation and task discrepancy) and promote the primary few-shot learning. Finally, we conduct extensive experiments on six benchmark datasets to validate the effectiveness and efficiency of TRGM. Experimental results show that our model outperforms several strong baselines and achieves the new state-of-the-art.

Fengzhao Shi, Yanan Cao, Yanmin Shang, Yuchen Zhou,Chuan Zhou,Jia Wu

Abstract: In the fraud graph, fraudsters often interact with a large number of benign entities to hide themselves. So, there are not only the homophilic connections formed by the same label nodes (similar nodes), but also the heterophilic connections formed by the different label nodes (dissimilar nodes). However, the existing GNN-based fraud detection methods just enhance the homophily in fraud graph and use the low-pass filter to retain the commonality of node features among the neighbors, which inevitably ignore the difference among neighbor of heterophilic connections. To address this problem, we propose a Graph Neural Network-based Fraud Detector with Homophilic and Heterophilic Interactions (H2-FDetector for short). Firstly, we identify the homophilic and heterophilic connections with the supervision of labeled nodes. Next, we design a new information aggregation strategy to make the homophilic connections propagate similar information and the heterophilic connections propagate difference information. Finally, a prototype prior is introduced to guide the identification of fraudsters. Extensive experiments on two real public benchmark fraud detection tasks demonstrate that our method apparently outperforms state-of-the-art baselines.

Ren Li, Yanan Cao, Qiannan Zhu, Guanqun Bi, Fang Fang, Yi Liu, Qian Li

Abstract: Knowledge Graph Embedding (KGE) aims to learn representations for entities and relations. Most KGE models have gained great success, especially on extrapolation scenarios. Specifically, given an unseen triple (h, r, t), a trained model can still correctly predict t from (h, r, ?), or h from (?, r, t), such extrapolation ability is impressive. However, most existing KGE works focus on the design of delicate triple modeling function, which mainly tell us how to measure the plausibility of observed triples, but we have limited understanding of why the methods can extrapolate to unseen data, and what are the important factors to help KGE extrapolate. Therefore in this work, we attempt to, from a data relevant view, study KGE extrapolation of two problems: 1. How does KGE extrapolate to unseen data? 2. How to design the KGE model with better extrapolation ability? For the problem 1, we first discuss the impact factors for extrapolation and from relation, entity and triple level respectively, propose three Semantic Evidences (SEs), which can be observed from training set and provide important semantic information for extrapolation to unseen data. Then we verify the effectiveness of SEs through extensive experiments on several typical KGE methods, and demonstrate that SEs serve as an important role for understanding the extrapolation ability of KGE. For the problem 2, to make better use of the SE information for more extrapolative knowledge representation, we propose a novel GNN-based KGE model, called Semantic Evidence aware Graph Neural Network (SE-GNN). Finally, through extensive experiments on FB15k-237 and WN18RR datasets, we show that SE-GNN achieves state-of-the-art performance on Knowledge Graph Completion task and perform a better extrapolation ability.

Xiaoxue Li, Yanan Cao*, Yanmin Shang, Yangxi Li, Qian Li, Guandong Xu

Abstract: User identity linkage is a task of recognizing the identities of the same user across different social networks (SN). Previous works tackle this problem via estimating the pairwise similarity between identities from different SN, predicting the label of identity pairs or selecting the most relevant identity pair based on the similarity scores. However, most of these methods ignore the results of previously matched identities, which could contribute to the linkage in following matching steps. To address this problem, we convert user identity linkage into a sequence decision problem and propose a reinforcement learning model to optimize the linkage strategy from the global perspective. Our method makes full use of both the social network structure and the history matched identities, and explores the long-term influence of current matching on subsequent decisions. We conduct experiments on different types of datasets, the results show that our method achieves better performance than other state-of-the-art methods.

Xiaoxue Li, Yanan Cao*, Yanmin Shang, Yangxi Li, Yanbing Liu, Jianlong Tan

Abstract: Anchor Link Prediction (ALP) across heterogeneous networks plays a pivotal role in inter-network applications. The difficulty of anchor link prediction in heterogeneous networks lies in how to consider the factors affecting nodes alignment comprehensively. In recent years, predicting anchor links based on network embedding has become the main trend. For heterogeneous networks, previous anchor link prediction methods first integrate various types of nodes associated with a user node to obtain a fusion embedding vector from global perspective, and then predict anchor links based on the similarity between fusion vectors corresponding with different user nodes. However, the fusion vector ignores effects of the local type information on user nodes alignment. To address the challenge, we propose a novel type-aware anchor link prediction across heterogeneous networks (TALP), which models the effect of type information and fusion information on user nodes alignment from local and global perspective simultaneously. TALP can solve the network embedding and type-aware alignment under a unified optimization framework based on a two-layer graph attention architecture. Through extensive experiments on real heterogeneous network datasets, we demonstrate that TALP significantly outperforms the state-of-the-art methods.