Sparse General Matrix Multiplication (SpMM) is a fundamental kernel in iterative scientific computing and graph neural network inference. On single-GPU platforms, iterative SpMM suffers from two major performance bottlenecks: excessive memory traffic caused by repeated loading of Krylov basis vectors, and low computational efficiency due to irregular sparsity patterns. To address these challenges, this article proposes BLR-Krylov , an iterative SpMM optimization framework that integrates Communication-Avoiding Krylov (CA-Krylov) methods with a Block Low-Rank Sparse (BLR-Sparse) matrix format. BLR-Krylov first preprocesses sparse matrices into the BLR-Sparse format, where low-rank blocks are compressed through basis vector decomposition to reduce storage and arithmetic overhead. It then employs GPU-optimized matrix power kernels to batch-compute Krylov subspace basis vectors, effectively minimizing global memory accesses across iterations. Experimental results on multiple GPU architectures demonstrate that BLR-Krylov achieves a 4.0–6.0× speedup over existing communication-avoiding approaches in iterative workloads, increases global memory bandwidth utilization to 82%–87%, and improves tensor core utilization to 85%–92%, while maintaining numerical accuracy within 1 × 10 -6 . To facilitate reproducibility, partial source code is publicly available at https://github.com/19547035579zz-tech/BLR-Krylov.

Affective events are events that are typically associated with a positive or negative emotional state. For example, get food is a desirable event, while suffer from asthma is an undesirable event. Identifying affective events and their polarities is essential for many artificial intelligence (AI) tasks. Since sentiments toward an affective event is often implicit, affective event acquisition and classification have always been two great challenges. Encounter events are a special type of affective events. The goal of this article is to automatically acquire and classify stereotypically positive and negative encounter events in Chinese. First of all, we collect a comprehensive list of Chinese encounter event indicators based on the literature related to Chinese encounter events. Then, we automatically extract candidate encounter events from a text corpus using these indicators as keywords. In order to acquire high-quality encounter events, we design an event filtering scheme and generate a novel stopword list. Finally, we classify these obtained encounter events exploiting the indicators. Experimental results show that the proposed approaches to acquire and classify Chinese encounter events perform promisingly well. One key advantage of the method is that no labelling and training of data is required. The automatically constructed common affective event knowledge base contains more than 38,000 encounter events annotated with affective polarity labels (negative or positive), which is currently the only resource of common affective events in Chinese language. Related resources including encounter event indicators and encounter events will be released after the publication of this article.

Unsupervised salient object detection is an important task in many real-world scenarios where pixel-wise label information is of scarce availability. Despite its significance, this problem remains rarely explored, with a few works that consider unsupervised salient object detection methods based on the fused graph from the sum fusion of multiple deep feature similarity matrices. However, these methods ignore the interrelation of the low-level feature similarity matrices and the high-level semantic similarity matrice, which degrades the quality of the fused graph. In this article, we propose a semantic-consistency-guided multi-graph fusion learning algorithm for unsupervised saliency detection, where the consistency and inconsistency between multiple low-level feature similarity matrices and the high-level semantic similarity matrice are explored to promote the robustness and quality of the fused graph. In the first stage, a semantic-consistency-guided multi-graph fusion learning method is proposed to exploit consistency and inconsistency of multiple low-level deep features and the high-level semantic feature. The semantic-consistency-guided similarity matrices are computed for preliminary saliency ranking. In the following saliency refinement stage, the semantic-enhanced similarity matrices are built by the cross diffusion to fuse the multiple low-level deep features and the high semantic deep feature. Based on the semantic-enhanced similarity matrices, the refinement saliency maps are calculated in a semantic-enhanced cellular automata manner. Furthermore, the final ensemble stage of the large margin semi-supervised classification views the preliminary ranking results and refinement results as features, adopts the large margin graphs for saliency ensemble. Extensive evaluations over four benchmark datasets show that the proposed unsupervised method performs favorably against the state-of-the-art approaches and is competitive with some supervised deep learning-based methods.