The prevalent class imbalance in real-world graphs significantly affects the performance of Graph Neural Networks (GNNs). Existing methods for analyzing graph imbalance ignore the influence of minority nodes during the dynamic model training process, resulting in performance limitations. In this paper, we focus on minority class information during model training, identifying and defining the minority class forgetting phenomenon that exists in graph imbalanced method training processes. To address this issue, we propose Graph Imbalance Experience Replay(GIER) framework. On one hand, the method enhances the model's ability to mine minority node information in historical data, thereby achieving feature completion for minority class nodes. On the other hand, the proposed short-term confidence mechanism allows the model to adaptively calibrate the topological relationships in high-confidence nodes, thereby mitigating the model's tendency to propagate erroneous information about minority classes during training. GIER is a unified framework consisting of two synergistic components: Long-term Subgraph Memory (LSM) constructs multi-period feature-representative subgraphs to address distribution imbalance, and Short-term Confidence Calibration (SCC) dynamically reconstructs graph topology through degree-aware node selection and confidence-based filtering to address topological imbalance. The extensive experimental results demonstrate that GIER effectively improves the classification performance of GNNs on imbalanced graphs, achieving up to a 3.44% improvement in BAcc over the state-of-the-art, and is particularly effective in extreme scenarios with very small minority classes.

The limited availability of high-quality training data poses a persistent challenge for synthetic aperture radar (SAR) target classification. Existing data augmentation methods mainly adopt a simplistic application of GAN-based style transfer techniques to directly synthesize pseudo-SAR images from optical images. However, our in-depth analysis of this cross-modal conversion reveals that such straightforward strategies primarily focus on transferring high-level semantic information (e.g., target shapes), thus failing to adequately capture the essential low-level features unique to SAR imagery (e.g., scattering textures). To address this inherent trade-off between high-level semantic preservation and low-level feature authenticity, we propose a Hierarchical Feature-Constrained GAN (HiFC-GAN) tailored for optical-to-SAR style transfer. Specifically, HiFC-GAN enhances the representation of low-level SAR features by introducing local texture contrast constraints at shallow layers, while introducing explicit feature mapping constraints at deeper layers to maintain high-level semantic consistency throughout the reconstruction process. Experimental results demonstrate that HiFC-GAN significantly outperforms existing GAN-based techniques in image generation quality, particularly improving the low-level feature authenticity of pseudo-SAR images. Moreover, the generated pseudo-SAR images further improve the performance of downstream target classification tasks, yielding accuracy gains ranging from 3.56% to 5.90% on average with mainstream CNN-based models.

Temporal Knowledge Graphs (TKGs) serve as crucial tools for representing dynamic changes in the real world. Extrapolation reasoning within TKGs aims to predict entirely unknown future facts based on limited historical data, offering considerable practical value across various fields. However, existing methods generally focus on the recurrence and periodicity of historical facts, while overlooking the dynamic interactions associated with future facts. Moreover, these methods fail to capture historical evolutionary patterns, which grow increasingly complex as historical data accumulates. To this end, we propose TCPN, a novel Temporal Pyramidal Recurrent Network with contrastive learning for TKG extrapolation reasoning. Specifically, TCPN leverages a temporal pyramidal recurrent network to capture historical dependencies across multiple temporal scales, thereby refining temporal feature representations over extended time spans. Furthermore, TCPN seamlessly integrates contrastive learning to effectively align historical information with query semantics relevant to future facts. Lastly, we incorporate an adaptive time-aware mechanism, which uniformly models long-short term dependencies in time series with different granularities, explicitly fusing temporal feature information. Extensive experiments on four widely used TKG datasets show that TCPN significantly outperforms state-of-the-art methods across all metrics.

Extrapolation reasoning on Temporal Knowledge Graphs (TKGs) plays a pivotal role in various systems, including retrieval, recommendation, and Q&A. Traditional TKG reasoning methods tend to emphasize modeling the local and global features of facts, often overlooking the alignment with query semantics. Crucially, these methods are challenging in generating explicit reasoning paths. To address these gaps, we propose an innovative framework (LogiQ) for extrapolation reasoning on TKGs, steered by temporal logic rules and queries. Specifically, LogiQ incorporates temporal logic rules and implements a rule-guided reward mechanism, directing Reinforcement Learning (RL) agents toward actions more aligned with rules. Additionally, LogiQ merges temporal queries with neighbor aggregation, ensuring that candidate actions not only encapsulate neighboring factual data but also embody query semantics. This dual focus enables LogiQ to guide actions to find reasoning paths in limited steps strategically. Extensive experiments on four real-world TKG datasets demonstrate the superior performance of LogiQ across all metrics compared to existing state-of-the-art models.

Session-based recommendation (SBR) provides personalized recommendations by analyzing the interactions of anonymous session users. Recent approaches based on graph neural networks (GNNs) focus on pairwise relations to infer potential user preferences. However, real-world user interactions are often complex, involving high-order relations that GNNs may not fully capture. Hypergraphs can naturally model high-order relations, while having untapped potential in SBR. In this paper, we propose a hypergraph convolutional network (MoHyNet) based on hypergraph motifs to improve SBR. Specifically, we construct a hypergraph convolution to extract high-order relations among users, thereby reducing the impact of intra-session irrelevant information on recommendations. Additionally, we introduce hypergraph motifs to characterize users’ behavioral patterns, thus enhancing inter-session information mining. Besides, we incorporate contrastive learning to strengthen the representation of the current session. Extensive experiments on multiple real-world datasets demonstrate the superiority of our proposed model over state-of-the-art approaches.

Class imbalance with node quantity or node topology challenges graph node classification in real-world, such as fake user identification and fraud finance detection in social networks. Current studies mitigate the detrimental effects of class imbalance on graph neural networks through data augmentation and weight adjustment. However, these methods generate unreliable data and lack global generalizability, which results in poor classification performance for class-imbalanced nodes. To tackle the above issues, we propose a Graph Diffusion Enhanced Hypergraph method (GraphDEH), which introduces a hypergraph to capture high-order data correlations, strengthening semantic relational space of nodes. Specifically, we design a generator to constructs reliable hyperedges, which mitigate insufficient learning of minority classes caused by node quantity imbalance. Furthermore, we design an enhancer that employs graph diffusion to evaluate global influence among nodes, which increases homogeneous information in neighborhoods of minority-class nodes for node topology imbalance. GraphDEH outperforms multiple benchmark methods in class-imbalanced node classification across four benchmark datasets.

Sharding is a promising technique for increasing a blockchain system’s throughput by enabling parallel transaction processing. The main challenge of state sharding lies in ensuring the atomicity verification of cross-sharding transactions, which results in double communication overhead and increases the transaction’s confirmation time. Previous research has primarily focused on developing cross-shard protocols for the fast and reliable validation of transactions involving multiple shards. These studies typically generate a large number of cross-shard transactions because they primarily use simple address mapping for state sharding, that is, the prefix/suffix of the account address. In this article, we propose a state sharding scheme via density-based partitioning of the account-transaction graph. In order to reduce cross-shard transactions, the scheme groups correlated accounts into the same shard by generating the densest subgraphs, as the graph density describes the correlation among accounts, i.e., how often transactions have occurred among accounts. We formulate the graph density-based state sharding problem, with the goal of maximizing the average density across all shards under the workload constraint. We prove the NP-completeness of the problem. To reduce the complexity of finding the densest subgraph, we propose the pruning-based algorithm that reduces the search space by pre-pruning some invalid edges based on the concept of core number. We also extend the linear deterministic greedy algorithm and PageRank algorithm to handle new transactions in the dynamic scenario. We conduct extensive experiments using real transaction data from Ethereum. The experimental results demonstrate a strong correlation between the shard density and the number of cross-shard transactions, and the pruning-based algorithm can reduce the running time by an order of magnitude.

Document-level event causality identification (DECI) aims to discern the causal relationship between two event mentions within a document, which is a more challenging task compared to sentence-level event causality identification. However, the conventional DECI model applies the same method to all pairs of events without considering the distinctions and connections between intra-sentence and inter-sentence event pairs. This paper introduces a novel Separate and Integrate Different Level reasoning model (SIDL) to extract and fuse text features of events. Specifically, we separately encode intra-sentence and inter-sentence event pairs and leverage attention from pre-trained language models to identify the relevant context for aiding in the determination of causality. Additionally, we propose a causal reasoning strategy that leverages intra-sentence causal relationships to assist in the inference of inter-sentence causal relationships. We combine different features of event pairs by discriminating the characters and effects of intra-sentence and inter-sentence reasoning. Experimental results show that our approach achieves superior performance compared to previous models in two benchmarks EventStoryLine and CausalTimeBank.

Tingxuan Chen, Jun Long, Liu Yang, Zidong Wang, Yongheng Wang, Xiongnan Jin. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Temporal Knowledge Graphs (TKGs) serve as indispensable tools for dynamic facts storage and reasoning. However, predicting future facts in TKGs presents a formidable challenge due to the unknowable nature of future facts. Existing temporal reasoning models depend on fact recurrence and periodicity, leading to information degradation over prolonged temporal evolution. In particular, the occurrence of one fact may influence the likelihood of another. To this end, we propose THCN, a novel Temporal Causal Convolutional Network based on Hawkes processes, designed for temporal reasoning under the extrapolation setting. Specifically, THCN harnesses a temporal causal convolutional network with dilated factors to capture historical dependencies among facts spanning diverse time intervals. Then, we construct a conditional intensity function based on Hawkes processes for fitting the likelihood of fact occurrence. Importantly, THCN pioneers a dual-level dynamic modeling mechanism, enabling the simultaneous capture of the collective features of nodes and the individual characteristics of facts. Extensive experiments on six real-world TKG datasets demonstrate our method significantly outperforms the state-of-the-art across all four evaluation metrics, indicating that THCN is more applicable for extrapolation reasoning in TKGs.

Surgical phase segmentation plays an important role in computer-assisted surgery systems, aiming to recognize what step or what action is operating in the video frame. Existing methods focus on improving the accuracy and precision of video segmentation, but ignore semantic consistency and temporal continuity of video frames in the intra-phase, which is necessary to apply in realistic computer-assisted equipment. Meanwhile, recent works almost extract long-term dependencies by Temporal Convolutional Network, but we heed high layers in TCN lose fine-grained information for detecting surgical steps and further affect phase segmentation task. To address these problems, we propose a Surgical Phase Segmentation Framework (SPHASE) which contains a multimodal feature fusion process and follows a multi-branch predictor. Moreover, we design a multimodal feature fusion mechanism when aggregate optical flow feature and I3D feature. The extensive experiments on AutoLaparo, Cholec80, and M2CAI2016 datasets demonstrate our method outperforms the state-of-the-art method by a large margin, especially in the JACC metric, which means SPHASE is more applicable in the surgical operating room.