In collaborative inference scenarios, semantic communication replaces raw data transmission by conveying task-oriented semantic features to improve bandwidth efficiency. However, under noisy wireless channels, the combined effects of semantic compression distortion and channel noise lead to severe information loss, resulting in degraded inference accuracy. To address this issue, this paper proposes a Multi-scale Semantic Compression Collaborative Inference (MSCCI) framework that achieves efficient, stable inference performance under high compression ratios and elevated noise levels. Specifically, a UNet-based encoder extracts multi-scale semantic features on an IoT device. These features are then integrated into a unified stream using a novel semantic fusion compression strategy, thereby substantially reducing communication overhead. The edge server decoder decompresses features and recovers image semantics via progressive upsampling and multi-scale semantic restoration. For noisy wireless channels, the framework incorporates Squeeze-and-Excite (SE) attention for dynamic feature channel weighting and residual connections for enhanced low-SNR robustness. Experimental results demonstrate that our collaborative inference framework for semantic communication outperforms state-of-the-art algorithms, and the approach’s effectiveness and robustness are verified across various channel conditions.

Ming potential patterns from patient data are usually treated as a high‐dimensional data clustering problem. Evolutionary multiobjective clustering algorithms with feature selection (FS) are widely used to handle this problem. Among the existing algorithms, FS can be performed either before or during the clustering process. However, research on performing FS at both stages (hybrid FS), which can yield robust and credible clustering results, is still in its infancy. This paper introduces an improved high‐dimensional patient data clustering algorithm with hybrid FS called variable dimensional multiobjective lifetime constrained quantum PSO with reinforcement learning (VLQPSOR). VLQPSOR consists of two main independent stages. In the first stage, a dimensionality reduction ensemble strategy is developed before clustering to reduce the patient dataset’s dimensionality, resulting in subdatasets of varying dimensions. In the second stage, an improved multiobjective QPSO clustering algorithm is proposed to simultaneously conduct dimensionality reduction and clustering. To accomplish this, several strategies are employed. Firstly, the variable dimensional lifetime constrained particle learning strategy, the continuous‐to‐binary encoding transformation strategy, and multiple external archives elite learning strategy are introduced to further reduce the dimensionality of the subdatasets and mitigate the risk of QPSO getting trapped in local optima. Secondly, an improved reinforcement learning–based clustering method selection strategy is proposed to adaptively select the optimal classical clustering algorithm. Experimental results demonstrate that VLQPSOR outperforms five representative comparative algorithms across four validity indexes and clustering partitions for most patient datasets. Ablation experiments confirm the effectiveness of the proposed strategies in enhancing the performance of QPSO.

The extraction of water bodies in densely built-up areas is often surrounded by complex background environments, with shadow regions caused by high-rise buildings in high-resolution remote sensing images being one of the most serious challenges in water body recognition research. Convolutional neural networks (CNNs) have significant research value in capturing spatial structures in images. Compared to other deep learning models, CNNs require fewer parameters and computational resources while effectively extracting local features, achieving satisfactory segmentation performance. However, in the existing high-resolution remote sensing images, the color and shape of shadow regions exhibit high similarity to water bodies. The use of single-scale structures and shared weights severely hinders the CNN model’s ability to accurately distinguish and extract water bodies under strong interference conditions. To address these issues, this letter proposes a lightweight CNN method (MSAttU-Net) to overcome the interference of urban shadow regions in large-scale high-resolution remote sensing images. First, a lightweight multibranch deep convolution module is constructed to expand the model’s receptive field and enhance its feature extraction capability. Second, a parallel attention mechanism module (P-Att) is introduced, consisting of parallel spatial and channel attention mechanisms. These are designed to improve the focus on water boundary and shape features as well as corresponding spectral information, reducing the interference from shadows and other noise in the images. Finally, a$1\times 1$convolution layer is used to generate pixel-level classification results. The proposed method’s performance is compared with that of contemporary CNN networks, including U-Net, Seg-Net, Res-Net, and Dense-Net, popular CNN networks, such as DeeplabV3+, HR-Net, and SegFormer, and the state-of-the-art (SOTA) networks, such as TransUnet, SwinUnet, and DS-TransUnet. The results indicate that the proposed model not only maintains a lightweight and efficient structure but also demonstrates exceptional capability in water body recognition and effectively overcoming shadow region interference.

Owing to their powerful semantic reasoning capabilities, Large Language Models (LLMs) have been effectively utilized as recommenders, achieving impressive performance. However, the high inference latency of LLMs significantly restricts their practical deployment. To address this issue, this work investigates knowledge distillation from cumbersome LLM-based recommendation models to lightweight conventional sequential models. It encounters three challenges: 1) the teacher’s knowledge may not always be reliable; 2) the capacity gap between the teacher and student makes it difficult for the student to assimilate the teacher’s knowledge; 3) divergence in semantic space poses a challenge to distill the knowledge from embeddings.

Reinforcement Learning (RL)-Based Recommender Systems (RSs) have gained rising attention for their potential to enhance long-term user engagement. However, research in this field faces challenges, including the lack of user-friendly frameworks, inconsistent evaluation metrics, and difficulties in reproducing existing studies. To tackle these issues, we introduce EasyRL4Rec, an easy-to-use code library designed specifically for RL-based RSs. This library provides lightweight and diverse RL environments based on five public datasets and includes core modules with rich options, simplifying model development. It provides unified evaluation standards focusing on long-term outcomes and offers tailored designs for state modeling and action representation for recommendation scenarios. Furthermore, we share our findings from insightful experiments with current methods. EasyRL4Rec seeks to facilitate the model development and experimental process in the domain of RL-based RSs. The library is available for public use.

Abstract Clustering divides objects into groups based on similarity. However, traditional clustering approaches are plagued by their difficulty in dealing with data with complex structure and high dimensionality, as well as their inability in solving multi‐objective data clustering problems. To address these issues, an evolutionary state‐based novel multi‐objective periodic bacterial foraging optimization algorithm (ES‐NMPBFO) is proposed in this article. The algorithm is designed to alleviate the high‐computing complexity of the standard bacterial foraging optimization (BFO) algorithm by introducing periodic BFO. Moreover, two learning strategies, global best individual (gbest) and personal historical best individual (pbest), are used in the chemotaxis operation to enhance the convergence speed and guide the bacteria to the optimum position. Two elimination‐dispersal operations are also proposed to prevent falling into local optima and improve the diversity of solutions. The proposed algorithm is compared with five other algorithms on six validity indexes in two data clustering cases comprising nine general benchmark datasets and four credit risk assessment datasets. The experimental results suggest that the proposed algorithm significantly outperforms the competing approaches. To further examine the effectiveness of the proposed strategies, two variants of ES‐NMPBFO were designed, and all three forms of ES‐NMPBFO were tested. The experimental results show that all of the proposed strategies are conducive to the improvement of solution quality, diversity and convergence.

This study aims to investigate the influencing factors of consumers’ repurchase intention in the context of online knowledge service. Drawing on information systems (IS) success model, social identity theory, and expectation-confirmation theory, we examine the roles of IS success factors, satisfaction, switching barrier, and brand awareness in influencing repurchase intention in the online knowledge service setting. In line with the three core components of online knowledge service (i.e., platform, product, and service provider), this research splits a holistic view of brand awareness into three facets, namely platform brand awareness, knowledge-product brand awareness, and service-provider brand awareness, and discusses their negative moderating roles. In particular, this study highlights the “double-edged sword” effect of brand awareness in the success of online knowledge service. Based on the survey of 301 users, the empirical results support most of our hypotheses. The theoretical contributions, practical implications, and limitations are also discussed.

New algebraic methods for constructing codes based on hyperplanes of two different dimensions in finite geometries are presented. The new construction methods result in a class of multistep majority-logic decodable codes and three classes of low-density parity-check (LDPC) codes. Decoding methods for the class of majority-logic decodable codes, and a class of codes that perform well with iterative decoding in spite of having many cycles of length 4 in their Tanner graphs, are presented. Most of the codes constructed can be either put in cyclic or quasi-cyclic form and hence their encoding can be implemented with linear shift registers.

This correspondence presents three algebraic methods for constructing low-density parity-check (LDPC) codes. These methods are based on the structural properties of finite geometries. The first method gives a class of Gallager codes and a class of complementary Gallager codes. The second method results in two classes of circulant-LDPC codes, one in cyclic form and the other in quasi-cyclic form. The third method is a two-step hybrid method. Codes in these classes have a wide range of rates and minimum distances, and they perform well with iterative decoding.