Graph Transformers (GTs), which integrate message passing and self-attention mechanisms simultaneously, have achieved promising empirical results in graph prediction tasks. However, the design of scalable and topology-aware node tokenization has lagged behind other modalities. This gap becomes critical as the quadratic complexity of full attention renders them impractical on large-scale graphs. Recently, Spiking Neural Networks (SNNs), as brain-inspired models, provided an energy-saving scheme to convert input intensity into discrete spike-based representations through event-driven spiking neurons. Inspired by these characteristics, we propose a linear-time Graph Transformer with Spiking Node Tokenization (GT-SNT) for node classification. By integrating multi-step feature propagation with SNNs, spiking node tokenization generates compact, locality-aware spike count embeddings as node tokens to avoid predefined codebooks and their utilization issues. The codebook guided self-attention leverages these tokens to perform node-to-token attention for linear-time global context aggregation. In experiments, we compare GT-SNT with other state-of-the-art baselines on node classification datasets ranging from small to large. Experimental results show that GT-SNT achieves comparable performances on most datasets and reaches up to 130× faster inference speed compared to other GTs.

Individual fairness (IF) in graph neural networks (GNNs), which emphasizes the need for similar individuals should receive similar outcomes from GNNs, has been a critical issue. Despite its importance, research in this area has been largely unexplored in terms of (1) a clear understanding of what induces individual unfairness in GNNs and (2) a comprehensive consideration of identifying similar individuals. To bridge these gaps, we conduct a preliminary analysis to explore the underlying reason for individual unfairness and observe correlations between IF andsimilarity consistency, a concept introduced to evaluate the discrepancy in identifying similar individuals based on graph structure versus node features. Inspired by our observations, we introduce two metrics to assess individual similarity from two distinct perspectives: topology fusion and feature fusion. Building upon these metrics, we proposeSimilarity-awareGNNs forIndividualFairness, namedSaGIF. The key insight behind SaGIF is the integration of individual similarities by independently learning similarity representations, leading to an improvement of IF in GNNs. Our experiments on several real-world datasets validate the effectiveness of our proposed metrics and SaGIF. Specifically, SaGIF consistently outperforms state-of-the-art IF methods while maintaining utility performance.

Large language models (LLMs) are widely adopted to generate synthetic datasets for various natural language processing (NLP) tasks, such as text classification and summarization. However, accurately measuring the diversity of these synthetic datasets—an aspect crucial for robust model performance—remains a significant challenge. In this paper, we introduce DCScore, a novel method for measuring synthetic dataset diversity from a classification perspective. Specifically, DCScore formulates diversity evaluation as a sample classification task, leveraging mutual relationships among samples. We further provide theoretical verification of the diversity-related axioms satisfied by DCScore, highlighting its role as a principled diversity evaluation method. Experimental results on synthetic datasets reveal that DCScore enjoys a stronger correlation with multiple diversity pseudo-truths of evaluated datasets, underscoring its effectiveness. Moreover, both empirical and theoretical evidence demonstrate that DCScore substantially reduces computational costs compared to existing methods. Code is available at: https://github.com/bluewhalelab/dcscore.

Real-world graphs are typically complex, exhibiting heterogeneity in the global structure, as well as strong heterophily within local neighborhoods. While a growing body of literature has revealed the limitations of graph neural networks (GNNs) in handling homogeneous graphs with heterophily, little work has been conducted on investigating the heterophily properties in the context of heterogeneous graphs. To bridge this research gap, we identify the heterophily in heterogeneous graphs using metapaths and propose two practical metrics to quantitatively describe the levels of heterophily. Our empirical investigations on real-world heterogeneous graphs have revealed that heterogeneous graph neural networks (HGNNs), which inherit many mechanisms from GNNs designed for homogeneous graphs, struggle to generalize to heterogeneous graphs with heterophily or low levels of homophily. To address the challenge, we present Hetero$^{2}$2Net, a heterophily-aware HGNN that incorporates masked metapath prediction and masked label prediction tasks to effectively and flexibly handle both homophilic and heterophilic heterogeneous graphs. We evaluate the performance of Hetero$^{2}$2Net on five real-world heterogeneous graph benchmarks with varying levels of heterophily. Experimental results demonstrate that Hetero$^{2}$2Net outperforms strong baselines in the semi-supervised node classification task. In particular, Hetero$^{2}$2Net scales to an industrial-scale commercial graph with 13 M nodes and 157 M edges, demonstrating its effectiveness in handling large and complex heterogeneous graphs.

While contrastive self-supervised learning has become the de-facto learning paradigm for graph neural networks, the pursuit of higher task accuracy requires a larger hidden dimensionality to learn informative and discriminative full-precision representations, raising concerns about computation, memory footprint, and energy consumption burden (largely overlooked) for real-world applications. This work explores a promising direction for graph contrastive learning (GCL) with spiking neural networks (SNNs), which leverage sparse and binary characteristics to learn more biologically plausible and compact representations. We propose SpikeGCL, a novel GCL framework to learn binarized 1-bit representations for graphs, making balanced trade-offs between efficiency and performance. We provide theoretical guarantees to demonstrate that SpikeGCL has comparable expressiveness with its full-precision counterparts. Experimental results demonstrate that, with nearly 32x representation storage compression, SpikeGCL is either comparable to or outperforms many fancy state-of-the-art supervised and self-supervised methods across several graph benchmarks.

Benefit from the widespread implement of data collection technologies, a large number of data services are emerged in various industries like finance and energy. The complex internal connectivity within these services can be naturally modeled as the graph-structured data, which facilitates services’ understanding based on graph mining. However, the heavy label reliance limits the development of traditional graph learning methods on data services. Graph contrastive learning (GCL), as a popular graph representation learning approach without relying on manual labels, has recently achieved a non-negligible effect. Thus, for alleviating above problems, we propose a simple yet effective method named FastGCL containing a better contrastive scheme which can be tailored to the characteristics of graph neural networks (e.g., neighborhood aggregation). Specifically, by constructing weighted-aggregated and non-aggregated neighborhood information as positive and negative samples respectively, FastGCL identifies the potential semantic information of data without disturbing the graph topology and node attributes, resulting in faster training and convergence speeds. Extensive experiments have been conducted on node classification and graph classification tasks, showing that FastGCL has competitive classification performance and significant training speedup compared to existing state-of-the-art methods.

Previous studies have confirmed microRNA (miRNA), small single-stranded non-coding RNA, participates in various biological processes and plays vital roles in many complex human diseases. Therefore, developing an efficient method to infer potential miRNA disease associations could greatly help understand operational mechanisms for diseases at the molecular level. However, during these early stages for miRNA disease prediction, traditional biological experiments are laborious and expensive. Therefore, this study proposes a novel method called AGAEMD (node-level Attention Graph Auto-Encoder to predict potential MiRNA Disease associations). We first create a heterogeneous matrix incorporating miRNA similarity, disease similarity, and known miRNA-disease associations. Then these matrixes are input into a node-level attention encoder-decoder network which utilizes low dimensional dense embeddings to represent nodes and calculate association scores. To verify the effectiveness of the proposed method, we conduct a series of experiments on two benchmark datasets (the Human MicroRNA Disease Database v2.0 and v3.2) and report the averages over 10 runs in comparison with several state-of-the-art methods. Experimental results have demonstrated the excellent performance of AGAEMD in comparison with other methods. Three important diseases (Colon Neoplasms, Lung Neoplasms, Lupus Vulgaris) were applied in case studies. The results comfirm the reliable predictive performance of AGAEMD.

The importance of microbe-drug associations (MDA) prediction is evidenced in research. Since traditional wet-lab experiments are both time-consuming and costly, computational methods are widely adopted. However, existing research has yet to consider the cold-start scenarios that commonly seen in clinical research and practices where confirmed MDA data are highly sparse. Therefore, we aim to contribute by developing two novel computational approaches, the GNAEMDA (Graph Normalized Auto-Encoder to predict MDA), and its variational extension (called VGNAEMDA), to provide effective and efficient solutions for well-annotated cases and cold-start scenarios. Multi-modal attribute graphs are constructed by collecting multiple features of microbes and drugs, and then input into a graph normalized convolutional network, where a $\ell _{2}$-normalization is introduced to avoid the norm-towards-zero tendency of isolated nodes in embedding space. Then the reconstructed graph output by the network is used to infer undiscovered MDA. The difference between the two proposed models lays in the way to generate the latent variables in network. To verify their effectiveness, we conduct a series of experiments on three benchmark datasets in comparison with six state-of-the-art methods. The comparison results indicate that both GNAEMDA and VGNAEMDA have strong prediction performances in all cases, especially in identifying associations for new microbes or drugs. In addition, we conduct case studies on two drugs and two microbes and find that more than 75% of the predicted associations have been reported in PubMed. The comprehensive experimental results validate the reliability of our models in accurately inferring potential MDA.