{"title":"Robust Semi-Supervised Anomaly Detection via Adaptive Boundary Exploration and Contrastive Representation Learning","authors":"Xinwei Wu, Cong Li, Xinsheng Ji, Weizhi Meng, Zilong Wang, Xingxing Liao","doi":"10.1016/j.eswa.2026.134976","DOIUrl":"10.1016/j.eswa.2026.134976","url":null,"abstract":"<div><div>The rapid evolution of the Internet of Things (IoT) has led to massive deployments of heterogeneous devices, generating large-scale, high-dimensional, and dynamic network traffic that poses significant challenges for anomaly detection. Semi-supervised anomaly detection methods leverage a small number of labeled anomalies to improve discrimination between normal and malicious traffic. However, under limited anomaly supervision, the available labeled anomalies provide only sparse information about the transition regions between normal and anomalous behaviors. Consequently, existing methods tend to construct decision boundaries passively from available training samples, leaving decision-critical regions insufficiently explored and making anomalies that closely resemble normal traffic difficult to distinguish. To address these limitations, we propose ABE-CRL, a semi-supervised anomaly detection framework that integrates Adaptive Boundary Exploration with Contrastive Representation Learning. ABE-CRL actively explores the normal–anomaly boundary while enhancing discriminative representation learning. Specifically, ABE-CRL introduces a metric-adaptive boundary exploration mechanism that assigns perturbation budgets according to the geometric relationships between labeled anomalies and a normal reference structure, thereby generating informative boundary-hard variants in decision-critical regions. These variants are further incorporated into a contrastive representation learning strategy that jointly organizes normal samples, labeled anomalies, and boundary-hard variants in the latent space, improving inter-class separability and reducing feature ambiguity near the decision boundary. Extensive experiments on nine benchmark intrusion detection datasets demonstrate that ABE-CRL outperforms state-of-the-art competitors by an average of 10.72% in AUC-PR and exhibits robust performance under different anomaly contamination levels and varying numbers of labeled anomalies.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"334 ","pages":"Article 134976"},"PeriodicalIF":9.4,"publicationDate":"2026-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149363229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"DPIR: Dynamic propagation inconsistency reasoning for explainable fake news detection with large vision-language models","authors":"Caixia Jing, Hang Gao, Yangtao Chen, Tiegang Gao","doi":"10.1016/j.eswa.2026.134733","DOIUrl":"10.1016/j.eswa.2026.134733","url":null,"abstract":"<div><div>With the rapid dissemination of multimodal content on social media, fake news detection faces increasingly complex challenges. Although existing methods can leverage textual and visual information for multimodal analysis, they often focus on static cross-modal consistency while overlooking the temporal-aware semantic divergence between news content and propagation evidence during information diffusion. To this end, this paper proposes an explainable fake news detection model based on <strong>D</strong>ynamic <strong>P</strong>ropagation <strong>I</strong>nconsistency <strong>R</strong>easoning (DPIR). Specifically, a temporal permeation strategy is introduced to dynamically fuse multimodal information and capture temporal-aware semantics within propagation sequences. Local semantic communities are leveraged to identify inconsistency patterns, and evidence-aware semantic aggregation is performed from a dynamic defense perspective. Finally, multi-source information is integrated to generate robust news representations. Extensive experiments conducted on three real-world datasets demonstrate that the proposed method outperforms the comparison methods in detection performance, achieving a 2.9% improvement in accuracy on the PHEME dataset, while providing reasonable and interpretable explanations for the detection results.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"334 ","pages":"Article 134733"},"PeriodicalIF":9.4,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149212388","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"MDFNet: a lightweight multi-scale dynamic fusion network for visual measurement of micro-defect in UAV power line inspection","authors":"Panpan Fang, Di Wu, Qin Wan, Yueping Xiao, Zhi Li","doi":"10.1016/j.eswa.2026.134737","DOIUrl":"10.1016/j.eswa.2026.134737","url":null,"abstract":"<div><div>The safe and stable operation of transmission lines is essential to the reliable operation of power grids. UAV-based visual inspection has emerged as an efficient paradigm for transmission-line inspection. However, UAV-based visual inspection still faces three major challenges: tiny components/defects, severe background clutter, and strict latency constraints due to limited onboard computing resources. In this paper, we propose MDFNet, a lightweight multi-scale fusion attention network for transmission-line inspection. Specifically, we first design a Multi-Receptive Field Aggregation (MRFA) module to aggregate multi-receptive-field responses, thus preserving fine structural cues for tiny targets. Then, a Global-Context Fusion Module (GCFM) is developed, which performs lightweight cross-scale context modeling via linear attention to enhance interactions among multi-level features. Finally, we design DAIM with parallel global and local attention to emphasize foreground components and suppress background interference. Extensive experiments on three benchmark datasets (Custom, InsPLAD, and CPLID) demonstrate that MDFNet consistently outperforms state-of-the-art methods, achieving significant gains of 13.5%, 8.1%, and 13.2% in small-object Average Precision (APS). With a compact footprint of 18.6 M parameters and an inference speed of 126.1 FPS, the proposed method offers an accurate and computationally efficient measurement solution for real-time power grid infrastructure monitoring</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"334 ","pages":"Article 134737"},"PeriodicalIF":9.4,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149201889","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Feifei Zhang, Kairui Bo, Siyuan Pan, Jie Chen, Hui Wang, Lei Xu, Chenchu Xu
{"title":"Interpretable Cine MR Image–Tabular Integration for Major Adverse Cardiovascular Event Prediction and Survival Analysis in Acute Myocardial Infarction","authors":"Feifei Zhang, Kairui Bo, Siyuan Pan, Jie Chen, Hui Wang, Lei Xu, Chenchu Xu","doi":"10.1016/j.eswa.2026.134685","DOIUrl":"10.1016/j.eswa.2026.134685","url":null,"abstract":"<div><div>Prediction of major adverse cardiovascular events (MACE) and survival analysis in acute myocardial infarction (AMI) patients are crucial for accurate identification of high-risk populations and personalized interventions. Existing methods mostly rely on electronic health records (EHR), but often do not fully exploit medical images that capture cardiac dynamics. Combining cine MR images with EHR data is expected to improve prediction performance and clinical practicality. However, due to problems such as modality heterogeneity, non-causal feature interference and class imbalance, the effective integration of the two still faces many challenges. This paper proposes a novel Cross-modal Alignment Causal Network (CACNet), that, for the first time, integrates cine MR images with EHR data for joint MACE prediction and survival analysis in AMI patients. The method consists of three key modules: The uncertainty-aware alignment module designs dynamic trust weights in contrastive learning to effectively regulate semantic uncertainty between modalities. The causality-driven interpretation module identifies myocardial motion regions associated with MACE risk, guiding the model to focus on key causally relevant features. The class-balanced reconstruction module alleviates the class imbalance problem by optimizing the decision boundary of the minority class. By integrating the above modules, CACNet achieves accurate prediction of cross-modal data. Systematic experiments on a real AMI dataset showed that CACNet achieves superior performance in both MACE prediction (AUROC: 88.53%, AUPRC: 69.14%) and survival analysis (C-index: 93.75%). The results validated the potential clinical value of CACNet in the early identification and personalized management of MACE in AMI patients.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"334 ","pages":"Article 134685"},"PeriodicalIF":9.4,"publicationDate":"2026-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149212602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"SPCA: Spatial Prompt with Cross-Attention for Zero-Shot Anomaly Detection","authors":"Yuqing Zhao, You Xiang, Ruiwei Xie","doi":"10.1016/j.eswa.2026.134505","DOIUrl":"10.1016/j.eswa.2026.134505","url":null,"abstract":"<div><div>Zero-shot anomaly detection (ZSAD) requires models trained on auxiliary data to detect anomalies in a target dataset without any in-category training samples. It is a challenging task due to significant cross-domain variations, where the appearance of objects and anomalies, such as defects in industrial products or lesions in medical images, can differ greatly. Recently, large pre-trained vision–language models (VLMs), such as CLIP, have shown strong zero-shot capabilities. However, existing prompt-based ZSAD methods are limited by coarse anomaly representations and an overemphasis on global semantic alignment, leading to weak discrimination and inaccurate localization. In this paper, we propose a novel framework, termed Spatial Prompt with Cross-Attention (SPCA), for robust ZSAD. The key idea is to enhance both semantic expressiveness and spatial modeling. Specifically, we introduce a Prompt Concatenation (PC) strategy to enrich anomaly prompts and a Cross-modal Semantic Fusion (CSF) module to improve image-level representations. We further propose Semantic Prompt Generation (SSG) to produce adaptive prompts for dense prediction, and Spatial Cross Attention (SCA) to strengthen fine-grained alignment for precise localization. Experiments on 17 datasets show that SPCA achieves superior zero-shot performance in both anomaly detection and segmentation across diverse domains.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"334 ","pages":"Article 134505"},"PeriodicalIF":9.4,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149212393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"A point cloud completion network integrating Mamba and transformer architectures","authors":"Weichao Wu , Yongyang Xu , Zhong Xie","doi":"10.1016/j.eswa.2025.130826","DOIUrl":"10.1016/j.eswa.2025.130826","url":null,"abstract":"<div><div>Point cloud completion aims to reconstruct complete structures from incomplete point clouds by extracting fine-grained local details and global features. Current state-of-the-art methods rely on Transformer architectures, which suffer from quadratic complexity, leading to high computational costs and trade-offs in resolution and feature extraction. To address this limitation, we propose a novel point cloud completion network that integrates the Mamba model, a state space framework with linear complexity, for feature extraction in the encoding phase. Our approach replaces the self-attention module with Mamba and introduces a multi-scale encoding network to enhance the extraction and fusion of features from incomplete point clouds. A cross-attention decoding module processes centre points and incomplete features to predict a complete point cloud. Experiments on synthetic and real-world datasets show that our method achieves comparable performance to existing state-of-the-art approaches on benchmark datasets, achieving an average CDL1 score of 6.50 on the PCN dataset. In addition, our method demonstrates superior accuracy when processing large-volume point cloud data, highlighting Mamba’s effectiveness in handling such challenges compared with Transformer-based models.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"313 ","pages":"Article 130826"},"PeriodicalIF":7.5,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Coherence-aware and snap-triggered: A novel mechanism for audio-visual cooperative tasks","authors":"Cunhan Guo, Heyan Huang, Ruiqi Hu, Danjie Han","doi":"10.1016/j.eswa.2026.131559","DOIUrl":"10.1016/j.eswa.2026.131559","url":null,"abstract":"<div><div>Audio-Visual Cooperative tasks underpin multimodal scene understanding and compel models to reconcile continuous temporal evolution with abrupt sensory transitions. We propose the Coherence-Aware and Snap-Triggered mechanism (CAST) mechanism, a plug-in temporal refinement layer without perturbing backbone parameters or demanding additional modalities. The Exponential Memory based Coherence-Aware module attenuates distant frame contributions through an exponentially decaying weight envelope, thereby preventing the persistent influence of obsolete disruptions. Complementarily, the Optical Flow based Snap-Triggered Module module registers instantaneous motion discontinuities and reallocates attention toward nascent events. Operating in concert, these modules yield a representation that remains coherent across smooth transitions yet responsive to sudden perturbations. Empirical evaluation across multiple AVC benchmarks demonstrates consistent superiority over established baselines, corroborating that CAST enhances temporal fidelity and, by extension, the reliability of downstream multimodal decisions.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"313 ","pages":"Article 131559"},"PeriodicalIF":7.5,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Breaking the low-cost barrier: a memory-augmented reactive navigation system for UAVs in cluttered indoor environments","authors":"Jiale Quan , Weijun Hu , Xianlong Ma , Gang Chen","doi":"10.1016/j.eswa.2026.131469","DOIUrl":"10.1016/j.eswa.2026.131469","url":null,"abstract":"<div><div>Achieving robust indoor autonomous flight for Unmanned Aerial Vehicles (UAVs) under strict hardware and computational constraints remains a formidable challenge. Conventional solutions relying on high-end sensors or global mapping are often inapplicable to resource-constrained micro-UAVs. In this paper, we propose a mapless integrated navigation framework aimed at achieving stable flight using a low-cost single-line 2D LiDAR. To address the limitations of sparse sensing, we propose a window-neighborhood-based denoising filtering algorithm and a velocity estimation-based motion distortion correction module. The system combines a risk-aware local planner and a short-sighted trajectory memory mechanism to navigate through cluttered spaces. The system operates in an O(N) loop with sub-millisecond latency. To overcome the local minima inherent in reactive planning, a deadlock escape layer is introduced, which formalizes navigation difficulty through trajectory entropy analysis, and generates recovery waypoints using discrete polar coordinate search. Validation through high-fidelity simulations and real-world experiments show that the system is capable of collision-free navigation at speeds up to 6 m/s, using low-cost sensors. This work provides an efficient solution for deploying intelligent aerial robots in perception-constrained indoor environments.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"313 ","pages":"Article 131469"},"PeriodicalIF":7.5,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xiujing Gao , Yongfeng Xie , Fanchao Lin , Chiwang Lin , Hongwu Huang , Ziru Wang
{"title":"IVYA-FMGRU: A frequency-domain context interaction model with bio-inspired optimization for significant wave height prediction","authors":"Xiujing Gao , Yongfeng Xie , Fanchao Lin , Chiwang Lin , Hongwu Huang , Ziru Wang","doi":"10.1016/j.eswa.2026.131534","DOIUrl":"10.1016/j.eswa.2026.131534","url":null,"abstract":"<div><div>Accurate prediction of significant wave heights is crucial for the safety of marine structures and ships. Traditional models struggle to capture the key frequency and periodic characteristics in wave height data. To address this issue, a novel Ivy Algorithm-Fast Fourier Transform Mogrifier Gated Recurrent Unit (IVYA-FMGRU) model is proposed, which integrates the gated recurrent unit (GRU) with the fast Fourier transform (FFT) and Mogrifier operations. The FFT extracts periodic features, the Mogrifier enhances the interaction between the GRU and frequency information, and the Ivy algorithm (IVYA), a bio-inspired optimization method, optimizes the model parameters. In addition, random forest (RF) is employed for feature selection. Experimental results show that the IVYA-FMGRU model achieves R<sup>2</sup> scores of 0.8505, 0.8683, and 0.8910 on datasets 46027, 46083, and 46084, respectively outperforming other baseline models. Furthermore, error statistical analysis across different wave height intervals confirms the model’s accuracy and stability within each interval, demonstrating its superior performance and generalization capability in wave height prediction.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"313 ","pages":"Article 131534"},"PeriodicalIF":7.5,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"An intelligent approach to maritime autonomous surface ship performance evaluation","authors":"Changyuan Chen, Chuanbo Duan, Yipeng Wang, Guiyang Zhang","doi":"10.1016/j.eswa.2026.131631","DOIUrl":"10.1016/j.eswa.2026.131631","url":null,"abstract":"<div><div>Maritime Autonomous Surface Ships (MASS) require reliable performance evaluation methods to ensure safe and efficient operation, yet existing research and regulations lack an integrated approach for automatic assessment of maneuvering, path planning, obstacle avoidance, and motion control performance. To fill this gap, this study proposes a comprehensive assessment tool that integrates assessment aspects, test scenarios, key performance indicators (KPIs), and evaluation criteria. Three assessment modules were developed: the Maneuvering Assessment Module (MAM), the Path planning and Obstacle avoidance Assessment Module (POAM), and the Motion Control Assessment Module (MCAM). The applicability of the proposed assessment tool is studied through simulations, including maneuvering tests under three water-depth conditions, comparative evaluation of ten path-planning algorithms, and an analysis of path-following control performance. The results indicate that maneuvering performance is poorer in shallow water (water depth-draft ratio <span><math><mrow><mi>h</mi><mo>/</mo><mi>T</mi></mrow></math></span>=1.4) compared with deep (<span><math><mrow><mi>h</mi><mo>/</mo><mi>T</mi></mrow></math></span>=10.0) and medium-deep water (<span><math><mrow><mi>h</mi><mo>/</mo><mi>T</mi></mrow></math></span>=2.0). Moreover, the proposed Tuned Fast Marching Square (TFMS) method generates safer and more cost-effective paths than Fast Marching Method (FMM) and FMS. These findings confirm that the assessment tool can provide quantitative and reproducible evaluations of MASS performance. The developed tool offers a practical platform for both researchers and practitioners, with potential extensions toward environmentally oriented (“green”) performance assessments in future work.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"313 ","pages":"Article 131631"},"PeriodicalIF":7.5,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}