Neurocomputing最新文献

{"title":"A dynamic hybrid expert framework with encoder–decoder interaction for robust image enhancement in train environment perception","authors":"Xin Liu ,&nbsp;Hongping Wang ,&nbsp;Linsen Song ,&nbsp;Yiwen Zhang ,&nbsp;Xiaoxu Zhang ,&nbsp;Chunbo Liu ,&nbsp;Xiao Shang ,&nbsp;Jingru Liu ,&nbsp;Yuanting Yang ,&nbsp;Xinming Zhang","doi":"10.1016/j.neucom.2025.130650","DOIUrl":"10.1016/j.neucom.2025.130650","url":null,"abstract":"<div><div>Train environment perception technology is one of the critical factors in ensuring safe train operations, particularly in challenging conditions such as foggy, rainy weather, and poorly lit environments like tunnels. The clarity of images directly influences the accuracy of obstacle detection and decision-making processes during train operation. However, existing image restoration methods are typically tailored to single scenarios, making them inadequate for the diverse and complex environmental variations encountered during train operations. Most of these methods lack specificity, rendering them ineffective in handling complex textures and fine details, resulting in suboptimal image quality under adverse conditions, often plagued by blurriness and noise interference. To address these challenges, we propose a dynamic hybrid expert image restoration framework specifically designed for train environment perception. This framework integrates multiple expert modules and a dynamic weight generation mechanism, enabling flexible adaptation to various environmental characteristics. Specifically, the framework comprises multiple expert modules, each focusing on distinct feature extraction tasks, thereby enhancing image clarity and detail restoration in challenging conditions such as foggy weather, low-light situations, and tunnels. The system dynamically generates weights based on the input image characteristics, allowing for the seamless integration of features extracted by each expert, which significantly improves image clarity and detail restoration. Additionally, the interaction between encoder–decoder attention mechanisms enhances the fusion of global and local information, ensuring robust image restoration in complex environments. Experimental results demonstrate that our method performs exceptionally well across various train operating conditions, particularly in foggy image enhancement and low-light image restoration in tunnels. Compared to existing methods, our approach achieves superior restoration quality and efficiency. Our method significantly enhances the image processing capabilities of train environment perception systems, providing a robust safeguard for safe train operations.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130650"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307227","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatial-frequency dual-domain Kolmogorov–Arnold networks for multimodal medical image fusion","authors":"Lewu Lin ,&nbsp;Jiaxin Xie ,&nbsp;Yingying Wang ,&nbsp;Jialing Huang ,&nbsp;Rongjin Zhuang ,&nbsp;Xiaotong Tu ,&nbsp;Xinghao Ding ,&nbsp;Na Shen ,&nbsp;Qing Lu","doi":"10.1016/j.neucom.2025.130661","DOIUrl":"10.1016/j.neucom.2025.130661","url":null,"abstract":"<div><div>Multimodal Medical Image Fusion (MMIF) can significantly enhance the efficiency and accuracy of clinical diagnosis and treatment by integrating medical images from different modalities into a single image with rich information. Recent advancements in Kolmogorov–Arnold Networks (KAN) have demonstrated significant potential in nonlinear fitting, owing to their ability to decompose complex multivariate functions into simpler univariate functions while maintaining high accuracy and interpretability. While most existing methods focus on developing increasingly complex architectures, addressing MMIF from a frequency analysis perspective and leveraging both spatial and frequency domains for interpretable and effective cross-modal fusion through KAN remains an underexplored frontier in prior research. To address this gap, we introduce Spatial-Frequency Dual-domain KAN (SFDKAN), a novel framework for MMIF. Initially, we apply a Hierarchical Wavelet Decomposition strategy to decompose the input modality into different frequency bands and introduce the powerful nonlinear mapping capability of KAN into the sub-bands of varying frequencies. This approach refines unimodal feature extraction and enhances the retention of high-frequency details and structural integrity. Next, we design a Spatial-Frequency Integration KAN (SFIKAN), leveraging complementary information from both spatial and frequency domains to facilitate effective cross-modality feature interaction and fusion. The Spatial KAN effectively focuses on critical regions in the fusion result, while ignoring irrelevant areas and suppressing redundant information. Meanwhile, the Frequency KAN overcomes the local limitations of the spatial domain, effectively handling long-range dependencies and enhancing global feature representation, thereby enabling more efficient cross-modality feature fusion. Extensive experiments on CI-MRI, PET-MRI, and SPECT-MRI datasets demonstrate the superiority of our method over state-of-the-art (SOTA) medical image fusion algorithms in both quantitative metrics and visual quality. The code will be available at <span><span>https://github.com/xiejiaaax/SFDKAN</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130661"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"fRAKI: k-space deep learning with offline data-universal and online scan-specific priors","authors":"Jing Pan ,&nbsp;Xiaohan Liu ,&nbsp;Yiming Liu ,&nbsp;Xuebin Sun ,&nbsp;Yanwei Pang","doi":"10.1016/j.neucom.2025.130663","DOIUrl":"10.1016/j.neucom.2025.130663","url":null,"abstract":"<div><div>Sampling a limited number of phase-encoding lines followed by estimating missing lines is an efficient method for shortening scan time of MRI. GeneRalized Autocalibarating Partial Parallel Acquisition (GRAPPA) is such a classical method and is widely used in clinical MRI. As a non-linear method, Robust Artificial-neural-networks for K-space Interpolation (RAKI) is a break-through of GRAPPA in the sense of much higher estimation accuracy. However, RAKI takes much longer estimation time because it requires online training a network for each receiving coil. To overcome the low-efficiency problem, we propose a fast version of RAKI (called fRAKI). fRAKI is roughly 26 times faster and can obtain much higher estimation accuracy compared with RAKI. The high efficiency of fRAKI is due to two properties: (1) A single network is shared to estimate missing lines of all the coils. (2) The online training of fRAKI can converge after a smaller number of iterations. Fast convergency is obtained by using a pre-trained model for initializing learnable parameters. High accuracy benefits from that the pre-train model contains data-universal prior and is also used as a sub-network of fRAKI so that the online training subnetwork can focus on learning scan-specific prior without the risk of overfitting the scan-specific data. Experimental results on the NYU fastMRI knee and brain datasets demonstrate the efficiency and accuracy of the proposed fRAKI.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130663"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144364897","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tuning multi-objective multi-population evolutionary models for high-dimensional problems: The case of the migration process","authors":"Juan Carlos Gómez-López ,&nbsp;Manuel Rodríguez-Álvarez ,&nbsp;Daniel Castillo-Secilla ,&nbsp;Jesús González","doi":"10.1016/j.neucom.2025.130631","DOIUrl":"10.1016/j.neucom.2025.130631","url":null,"abstract":"<div><div>Multi-objective multi-population evolutionary procedures have become one of the most outstanding metaheuristics for solving problems characterized by the <em>curse of dimensionality</em>. A critical aspect of these models is the migration process, defined as the exchange of individuals between subpopulations every few iterations or generations, which has typically been adjusted according to a set of guidelines proposed more than 20 years ago, when the capacity to deal with problems was significantly less than it is today. However, the constant increase in computational power has made it possible to tackle today’s complex real-world problems of great interest more plausibly, but with larger populations than before. Against this background, this paper aims to study whether these classical recommendations are still valid today, when both the magnitude of the problems and the size of the population have increased considerably, considering how this adjustment affects the performance of the procedure. In addition, the increase in the population size, coupled with the fact that multi-objective optimization is being addressed, has led to the development of a novel elitist probabilistic migration strategy that considers only the Pareto front. The results show some interesting and unexpected conclusions, in which other issues, such as the number of subpopulations or their size, should be considered when fitting multi-population models. Furthermore, some of the previously mentioned classical recommendations may not be well-suited for high-dimensional problems.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130631"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Neural network-based distributed adaptive fault-tolerant containment control","authors":"Ziying Fang ,&nbsp;Xiaojian Yi ,&nbsp;Tao Xu ,&nbsp;Xiaoguang Wang","doi":"10.1016/j.neucom.2025.130643","DOIUrl":"10.1016/j.neucom.2025.130643","url":null,"abstract":"<div><div>In practical applications, multi-agent systems (MASs) often face challenges arising from incomplete knowledge of system dynamics, and agent actuators may suffer from faults such as partial failures or biased inputs. This paper investigates the fault-tolerant containment control problem for nonlinear MASs subject to actuator faults and proposes a neural network-based control approach. The system model is assumed to involve unknown nonlinearities, and the follower agents may experience actuator faults. Neural networks are employed to approximate the unknown nonlinear dynamics, and adaptive parameters are introduced and updated online based on the system evolution. An adaptive distributed fault-tolerant control protocol is developed by integrating neural network approximations, adaptive parameter adjustments, and relative state errors between neighboring agents. By dynamically tuning the control effort through the adaptive parameters, the proposed protocol effectively compensates for system nonlinearities and ensures the achievement of the containment control objective, even in the presence of actuator faults. Simulation results are presented to demonstrate the effectiveness of the proposed control strategy.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130643"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interest transfer graph convolutional networks for multi-behavior recommendation","authors":"Minjie Fan,&nbsp;Yongquan Fan,&nbsp;Yajun Du,&nbsp;Xianyong Li","doi":"10.1016/j.neucom.2025.130659","DOIUrl":"10.1016/j.neucom.2025.130659","url":null,"abstract":"<div><div>Multi-behavior recommendation is an effective approach to address the problem of data sparsity. Graph-based multi-behavior recommendation is one of the most promising branches. Most research on multi-behavior recommendation uses Graph Convolutional Networks (GCNs) to model user features, as GCNs can capture high-order relationships and global features between nodes. However, the existing approaches suffer from multi-level user interests and noisy interactions. To address this issue, we propose a novel Interest Transfer Graph Convolutional Networks (ITGCN) for multi-behavior recommendation. Specifically, to model multi-level user interests, we designed a multi-level GCN by removing multi-layer aggregation operations to capture high-order relationships between nodes. In addition, to address the issue of noisy interactions, we propose a multi-behavior interest transfer method. This approach uses similarity-based comparison to reduce the impact of noisy interactions. It makes both target and auxiliary behaviors more robust to noise. At the same time, it transfers interests from auxiliary behaviors into the semantic space of the target behavior. Experiments on four datasets demonstrated the effectiveness of ITGCN.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130659"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144313568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"RoSiLC-RS:A Robust Similar Legal Case Recommender System Empowered by Large Language Model and Step-Back Prompting","authors":"Guohang Zeng ,&nbsp;George Tian ,&nbsp;Guangquan Zhang ,&nbsp;Jie Lu","doi":"10.1016/j.neucom.2025.130660","DOIUrl":"10.1016/j.neucom.2025.130660","url":null,"abstract":"<div><div>Legal case recommendation systems face significant challenges in the era of Large Language Models (LLMs). While LLMs offer unprecedented opportunities for understanding legal texts, they also introduce risks through AI-generated false legal content. Our survey reveals concerning gaps in public awareness: 41% of respondents incorrectly believe AI never generates false legal information, while only 6% understand potential legal liabilities. To address these issues, we propose RoSiLC-RS, a Robust Similar Legal Case Recommender System that guides LLMs to understand legal concepts at a higher abstraction level. Our system employs four key components: (1) abstraction processing to extract core legal elements, (2) semantic matching to identify similar case features, (3) LLM-powered explanation generation to provide detailed recommendation rationales, enhancing system explainability, and (4) a specialized detection module to identify and filter AI-generated false content. Comprehensive experiments on real-world legal datasets demonstrate that our method significantly outperforms traditional retrieval approaches in precision, relevance, explainability, and resistance to AI-generated content interference. This research provides both technological solutions and insights for the safe application of LLMs in legal domains.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130660"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144322534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learning referee evaluation and assessing action quality from coarse to fine in diving sport","authors":"Hong-Ming Qiu ,&nbsp;Hong-Bo Zhang ,&nbsp;Qing Lei ,&nbsp;Jing-Hua Liu ,&nbsp;Ji-Xiang Du","doi":"10.1016/j.neucom.2025.130664","DOIUrl":"10.1016/j.neucom.2025.130664","url":null,"abstract":"<div><div>Intelligently assessing the quality of athletic performances in sports scenarios remains a fascinating challenge in computer vision. However, unraveling the subtle distinctions between two similar actions in videos and mapping those video representations to quality scores remain significant obstacles. To address these challenges, this work redefines the paradigm of quality score estimation from traditional relative quality score prediction to relative referee score prediction. To make this shift, a cross-feature fusion module rooted in Transformer-based video representation is introduced, to improve pairwise video feature learning in the realm of action quality assessment. Then, a novel contrastive action parsing decoder module generates mid-level representations to effectively connect visual features with detailed quality scores. Both modules utilize cross-attention mechanisms; the former refines the pairwise video features to represent the differences between video pairs, while the latter updates the input queries corresponding to each referee’s evaluation. Finally, to achieve precise quality score estimation, we introduce a meticulous coarse-to-fine decision process, integrating a score classifier and offset regressor. After validation on challenging diving datasets, including MTL-AQA, FineDiving, and TASD-2, the experimental results show that the proposed approach demonstrates effectiveness and feasibility when compared with state-of-the-art methods.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130664"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Language-guided Alignment and Distillation for Source-free Domain Adaptation","authors":"Jiawen Peng,&nbsp;Jiaxin Chen,&nbsp;Rong Pan,&nbsp;Andy J. Ma","doi":"10.1016/j.neucom.2025.130501","DOIUrl":"10.1016/j.neucom.2025.130501","url":null,"abstract":"<div><div>Source-free domain adaptation (SFDA) is a practical problem in which a pre-trained source model is adapted to an unlabeled target domain without accessing the labeled source data. Although recent studies have successfully incorporated vision–language models (VLMs) like CLIP into SFDA frameworks, the performance of existing methods may be limited due to their reliance on coarse-grained class prompts, in which fine-grained textual knowledge has not been fully exploited. To overcome this limitation, we develop a novel framework of Language-guided Alignment and Distillation (LAD) by integrating visual features with fine-grained textual descriptions generated by pre-trained captioning models. Our method consists of two innovative designs, i.e., category-aware modality alignment (CMA) and language-guided knowledge distillation (LKD). CMA aligns cross-modal feature representations with a gating function to filter out high-confidence same-class samples from negatives to preserve intra-class similarity. LKD better adapts the vision encoder to the target domain through adaptive modality fusion and dual-level distillation guided by both visual and textual modalities. Extensive experiments on five benchmarks, including <em>both image and video recognition</em>, demonstrate that our method consistently outperforms the state of the arts for SFDA, e.g., +2.1% in Office-Home and +4.3% in UCF-HMDB.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130501"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144322532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MTR-MSE: Motion-Text Retrieval Method Based on Motion Semantics Expansion","authors":"Daoliang Xu ,&nbsp;Tianyou Zheng ,&nbsp;Yang Zhang ,&nbsp;Xiaodong Yang ,&nbsp;Weiwei Fu","doi":"10.1016/j.neucom.2025.130632","DOIUrl":"10.1016/j.neucom.2025.130632","url":null,"abstract":"<div><div>The motion-text cross-retrieval task aims to bridge the motion and text spaces, enabling mutual retrieval between motion and language. However, existing methods suffer from limited feature extraction due to both insufficient data and inadequate feature extraction techniques, which restrict retrieval accuracy and semantic richness. To address this, we propose a Motion-Text Retrieval Method Based on Motion Semantics Expansion (MTR-MSE). We design specialized motion and text encoders to create a comprehensive shared feature space. Furthermore, recognizing the limitations of overly simplistic textual descriptions in existing datasets, we enhance motion semantics using large language models to generate more detailed and varied descriptions, thereby improving motion understanding. Experimental results demonstrate that our method achieves state-of-the-art performance, validating its effectiveness in addressing the challenges of cross-modal motion-text retrieval.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"648 ","pages":"Article 130632"},"PeriodicalIF":5.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144312574","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}