Neurocomputing最新文献_第6页

Dynamic patch selection and dual-granularity alignment for cross-modal retrieval 跨模态检索的动态补丁选择和双粒度对齐

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-09 DOI: 10.1016/j.neucom.2026.132999

Zhenghui Luo, Min Meng, Jigang Wu

{"title":"Dynamic patch selection and dual-granularity alignment for cross-modal retrieval","authors":"Zhenghui Luo, Min Meng, Jigang Wu","doi":"10.1016/j.neucom.2026.132999","DOIUrl":"10.1016/j.neucom.2026.132999","url":null,"abstract":"<div><div>Cross-modal retrieval aims to establish semantic associations between heterogeneous modalities, among which image-text retrieval is a key application scenario that seeks to achieve efficient semantic alignment between images and texts. Existing approaches often rely on fixed patch selection strategies for fine-grained alignment. However, such static strategies struggle to adapt to complex scene variations. Moreover, fine-grained alignment methods tend to fall into local optima by overemphasizing local feature details while neglecting global semantic context. Such limitations significantly hinder both retrieval accuracy and generalization performance. To address these challenges, we propose a Dynamic Patch Selection and Dual-Granularity Alignment (DPSDGA) framework that jointly enhances global semantic consistency and local feature interactions for robust cross-modal alignment. Specifically, we introduce a dynamic sparse module that adaptively adjusts the number of retained visual patches based on scene complexity, effectively filtering redundant information while preserving critical semantic features. Furthermore, we design a dual-granularity alignment mechanism, which combines global contrastive learning with local fine-grained alignment to enhance semantic consistency across modalities. Extensive experiments on two benchmark datasets, Flickr30k and MS-COCO, demonstrate that our method significantly outperforms existing approaches in image-text retrieval.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 132999"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172689","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}

引用次数: 0

ASHSR: Enhancing query-based occupancy prediction via anti-occlusion sampling and hard sample reweighting ASHSR：通过抗遮挡采样和硬样本重加权增强基于查询的占用预测

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-11 DOI: 10.1016/j.neucom.2026.133007

Zhihao Li, Shanshan Zhang, Jian Yang

{"title":"ASHSR: Enhancing query-based occupancy prediction via anti-occlusion sampling and hard sample reweighting","authors":"Zhihao Li, Shanshan Zhang, Jian Yang","doi":"10.1016/j.neucom.2026.133007","DOIUrl":"10.1016/j.neucom.2026.133007","url":null,"abstract":"<div><div>3D occupancy prediction estimates the occupancy status in 3D space and is crucial for safe autonomous driving. Existing query-based methods improve efficiency but suffer from challenges such as poor quality of sampled features, semantic inconsistency of prediction results, and so on. To address these challenges, we propose ASHSR, an efficient occupancy prediction network driven by three innovative modules in the decoder. First, considering the limitations of the existing sampling approaches, we propose a view-aware Anti-occlusion Sampling Module (ASM), which significantly improves the quality of sampled features by adaptively avoiding meaningless sampling along the optical ray. Second, to resolve the ambiguity of the one-to-many prediction paradigm in single-query prediction, we design a Hard Sample Reweighting (HSR) module, which effectively improves the prediction purity and semantic consistency by assigning higher weights to the identified hard samples. Finally, to mitigate the cross-class data imbalance in driving scenes, we develop a hierarchical Coarse-to-Fine Supervision (CFS) mechanism to customize reasonable supervision signals across decoder layers, which significantly enhances the perception of rare categories. Extensive experiments on the Occ3D-nuScenes dataset demonstrate that our ASHSR significantly outperforms the existing state-of-the-art methods in terms of the RayIoU metric while maintaining superior inference efficiency (FPS).</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 133007"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172690","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}

引用次数: 0

Image-text driven style randomization for domain generalized semantic segmentation 面向领域广义语义分割的图像文本驱动风格随机化

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-04 DOI: 10.1016/j.neucom.2026.132953

Junho Lee , Jisu Yoon , Jisong Kim , Jun Won Choi

{"title":"Image-text driven style randomization for domain generalized semantic segmentation","authors":"Junho Lee , Jisu Yoon , Jisong Kim , Jun Won Choi","doi":"10.1016/j.neucom.2026.132953","DOIUrl":"10.1016/j.neucom.2026.132953","url":null,"abstract":"<div><div>Semantic segmentation models trained on source domains often fail to generalize to unseen domains due to domain shifts caused by varying environmental conditions. While existing approaches rely solely on text prompts for domain randomization, their generated styles often deviate from real-world distributions. To address this limitation, we propose a novel two-stage framework for <em>Domain Generalization in Semantic Segmentation</em> (DGSS). First, we introduce <em>Image-Prompt-driven Instance Normalization</em> (I-PIN), which leverages both style images and text prompts to optimize style parameters, achieving more accurate style representations compared to text-only approaches. Second, we present <em>Dual-Path Style-Invariant Feature Learning</em> (DSFL) which employs inter-style and intra-style consistency losses, ensuring consistent predictions across different styles while promoting feature alignment within semantic classes. Extensive experiments demonstrate that our approach consistently outperforms existing state-of-the-art methods across multiple challenging domains, effectively addressing the domain shift problem in semantic segmentation.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 132953"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172681","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}

引用次数: 0

Faithful explanation of semantic role labelling with dependency and constituency feature importance 语义角色标签与依赖和选区特征重要性的忠实解释

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-11 DOI: 10.1016/j.neucom.2026.132996

Hoai-Duc Tuan-Nguyen , Bac Le

{"title":"Faithful explanation of semantic role labelling with dependency and constituency feature importance","authors":"Hoai-Duc Tuan-Nguyen , Bac Le","doi":"10.1016/j.neucom.2026.132996","DOIUrl":"10.1016/j.neucom.2026.132996","url":null,"abstract":"<div><div>Syntactic features such as dependency and constituency are well known to improve Semantic Role Labeling (SRL), yet existing syntax-aware SRL models remain largely opaque. Related explainable NLP methods focus primarily on token-level representations and do not quantify how such relations contribute to individual SRL predictions. This limitation is particularly concerning in high-stakes domains such as biomedicine, where interpretability is essential for building trust in NLP-assisted analysis. In this work, we propose a post-hoc explanation framework that targets pairwise syntactic relations and provides a quantitative characterization of their influence on SRL predictions, measuring both the magnitude and polarity of their effects on argument span predictions. The framework employs a controlled representation perturbation that allows relation-specific analysis of importance, without modifying tokens or altering the model architecture. We further address the challenge of evaluating explanation quality without human-annotated importance labels by introducing a set of model-grounded diagnostic tests. These analyses assess whether the explanation scores systematically align with the model’s behavioral responses to syntactic perturbations, rather than relying on correlation with external judgments. To facilitate controlled experimentation, we also construct a lightweight syntax-aware SRL model via confidence-weighted task-vector merging, which avoids runtime parsing and annotated syntax at inference time. While auxiliary to the explanation framework, this model achieves competitive performance on biomedical text with reduced computational cost. Experiments on biomedical- and general-domain data demonstrate consistent behavioral patterns across domains, supporting the utility of the proposed explanations for analyzing how SRL models leverage syntactic structure.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 132996"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386327","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}

引用次数: 0

Multi-feature collaboration with spatial-frequency learning guided by vision foundation model for remote sensing image captioning 基于视觉基础模型的多特征协同与空间频率学习遥感图像字幕

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-06 DOI: 10.1016/j.neucom.2026.132936

Yijie Zhang, Jian Cheng, Ziying Xia, Siyu Liu, Changjian Deng, Zunni Zhu, Zichong Chen

{"title":"Multi-feature collaboration with spatial-frequency learning guided by vision foundation model for remote sensing image captioning","authors":"Yijie Zhang, Jian Cheng, Ziying Xia, Siyu Liu, Changjian Deng, Zunni Zhu, Zichong Chen","doi":"10.1016/j.neucom.2026.132936","DOIUrl":"10.1016/j.neucom.2026.132936","url":null,"abstract":"<div><div>Remote Sensing Image Captioning (RSIC) refers to the core task of utilizing technology to automatically generate natural language text, that accurately describes the types, spatial distribution, and complex semantic relationships of ground objects in remote sensing images. However, the spatial semantic relationships of ground objects in remote sensing images are complex and their scales are variable, making it challenging to achieve accurate descriptions. In this paper, we propose a multi-feature collaborative network (MFC-Net) with spatial-frequency learning guided by a vision foundation model (VFM). MFC-Net follows an encoder–decoder architecture: the encoder combines multi-feature representations in both spatial and frequency domains, while the decoder utilizes a transformer-based structure for caption generation. To better comprehend the unique geographical attributes and spatial relationships of ground objects, we leverage domain knowledge captured by the pre-trained Remote Sensing Vision Foundation Model (RS-VFM) to collaboratively guide the visual features generated by the CLIP image encoder and ResNet-50 branch through the Contextual Semantic Guidance Model (CSGM) and the Contextual Detail Guidance Model (CDGM). Furthermore, to better understand the multi-scale variations of ground objects, we transform multi-level features into the frequency domain using wavelet transform, efficiently extracting multi-scale features by partitioning them into frequency bands. Extensive experiments on two benchmark RSIC datasets demonstrate the superiority of our approach.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 132936"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386383","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}

引用次数: 0

Latent selective state-space models for partial differential equations via progressive learning 基于渐进式学习的偏微分方程潜在选择状态空间模型

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-11 DOI: 10.1016/j.neucom.2026.133000

A.Aoming Liang , B.Zhaoyang Mu , C.Pengxiao Lin

{"title":"Latent selective state-space models for partial differential equations via progressive learning","authors":"A.Aoming Liang , B.Zhaoyang Mu , C.Pengxiao Lin","doi":"10.1016/j.neucom.2026.133000","DOIUrl":"10.1016/j.neucom.2026.133000","url":null,"abstract":"<div><div>Partial Differential Equations (PDEs) play a fundamental role in modeling scientific and natural systems, such as fluid dynamics and weather forecasting. The integration of artificial intelligence into scientific computing has garnered significant attention in recent years. While many researchers focus on enhancing precision, the equally critical aspects of computational speed and training costs—particularly in long-term forecasting and control tasks—are often overlooked. This paper proposes MambaPDE, a novel approach that leverages latent vectors and state-space models to accelerate computations. To further improve accuracy, a progressive learning strategy is employed during the training phase. By comparing MambaPDE with multiple baseline methods across various datasets, this study demonstrates its effectiveness. Furthermore, our analysis shows that MambaPDE can effectively improve the condition number of the latent space, which is beneficial for the training stage. MambaPDE achieves competitive parameter efficiency, making it particularly well suited for applications requiring long-term forecasting, PDE control, and multi-modality simulation tasks.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 133000"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386330","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}

引用次数: 0

Diffbias: Harnessing diffusion models’ prediction bias for adversarial patch defense 扩散：利用扩散模型对对抗性斑块防御的预测偏差

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-09 DOI: 10.1016/j.neucom.2026.133009

Xudong Ye , Qi Zhang , Yapeng Wang , Xu Yang , Zuobin Ying , Jingzhang Sun , Qi Zhong , Xia Du

{"title":"Diffbias: Harnessing diffusion models’ prediction bias for adversarial patch defense","authors":"Xudong Ye , Qi Zhang , Yapeng Wang , Xu Yang , Zuobin Ying , Jingzhang Sun , Qi Zhong , Xia Du","doi":"10.1016/j.neucom.2026.133009","DOIUrl":"10.1016/j.neucom.2026.133009","url":null,"abstract":"<div><div>Adversarial patches pose a significant and real threat to deep neural networks, capable of inducing misclassification in realistic physical scenarios. Developing reliable and robust defense methods against these attacks is a critical application, and current research remains unsatisfactory. In this paper, we propose a novel framework that exploits the fact that unnatural perturbations introduced by adversarial patches can produce prediction biases significantly different from those of clean images during denoising. In the localization stage, our method focuses on the critical denoising steps through an adaptive temporal sampling strategy and introduces an energy metric that fuses kinetic and potential energy to quantify the degree of anomaly in the denoised trajectory. Furthermore, by combining this with the adaptive similarity weighting mechanism and the striding trajectory consistency analysis, our method effectively suppresses the interference of background noise, so as to achieve accurate locking of the patch area. In the restoration phase, the same diffusion model is applied to the patch region to restore the original visual content and integrity. This two-stage architecture shares a unified diffusion model, enabling the localization and inpainting processes to enhance the overall defense performance through information complementarity. Extensive experiments on the INRIA, COCO2017, and APRICOT datasets show that our approach achieves state-of-the-art detection performance under both digital and physical attack types without compromising the recognition accuracy of clean images.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 133009"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172693","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}

引用次数: 0

A novel dynamic graph generative adversarial network with edge-level differential privacy 一种具有边缘差分隐私的动态图生成对抗网络

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-10 DOI: 10.1016/j.neucom.2026.132926

Liya Ma , Siqi Sun , Chen Li , Lu Liu , Lihong Wang

{"title":"A novel dynamic graph generative adversarial network with edge-level differential privacy","authors":"Liya Ma , Siqi Sun , Chen Li , Lu Liu , Lihong Wang","doi":"10.1016/j.neucom.2026.132926","DOIUrl":"10.1016/j.neucom.2026.132926","url":null,"abstract":"<div><div>Most real-world graphs are dynamic and pose privacy risks when published and shared. Several researchers have used differential privacy to protect graph privacy. However, it is still challenging to maintain a continuous privacy–utility tradeoff when differential privacy methods are applied to dynamic graphs. To address this, we propose a novel Dynamic Graph Generative Adversarial Network with Edge-level Differential Privacy (<strong>DyGAN-EDP</strong>) model, which applies structural self-attention and an LSTM neural network to learn complex nonlinear patterns associated with graph structure evolution, which is beneficial for maintaining utility. To protect privacy, noise is introduced into the gradient of the generator during the training process to synthesize dynamic graphs, which can guarantee edge-level differential privacy. Moreover, a noise adjustment mechanism based on mutual information estimation is introduced, which provides stronger privacy protection when the graph structure changes significantly over time; at other times, it lowers the noise scale to ensure data utility. Experimental validations on three real-world dynamic graphs demonstrate the model’s ability to balance privacy and utility.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 132926"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172680","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}

引用次数: 0

Synchronization of discrete-time T-S fuzzy multi-layer networks under hybrid cyber attacks: An improved switching-like adaptive memory event-triggered mechanism 混合网络攻击下离散时间T-S模糊多层网络的同步：一种改进的类切换自适应记忆事件触发机制

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-11 DOI: 10.1016/j.neucom.2026.133054

Yunxiao Cao , Chuan Zhang , Xiwei Liu , Zi-Peng Wang

{"title":"Synchronization of discrete-time T-S fuzzy multi-layer networks under hybrid cyber attacks: An improved switching-like adaptive memory event-triggered mechanism","authors":"Yunxiao Cao , Chuan Zhang , Xiwei Liu , Zi-Peng Wang","doi":"10.1016/j.neucom.2026.133054","DOIUrl":"10.1016/j.neucom.2026.133054","url":null,"abstract":"<div><div>This paper investigates the synchronization control problem of a class of discrete-time T-S fuzzy multi-layer networks (MLNs) under simultaneous deception attacks (DAs) and denial of service (DoS) attacks. To alleviate the impact of limited network resources and hybrid cyber attacks, an improved adaptive memory event-triggered mechanism (ETM) with switching-like adaptive law and a fuzzy controller is designed. This mechanism can fully utilize historical error data and adaptively adjust the event-triggered threshold based on its long-term dynamic characteristics, thereby optimizing communication efficiency while ensuring control performance. On this basis, sufficient conditions for achieving intra-layer synchronization (ALS) and inter-layer synchronization (RLS) in discrete-time T-S fuzzy MLNs are derived by constructing appropriate Lyapunov functions and combining them with linear matrix inequality methods. Finally, the feasibility and effectiveness of the proposed method are verified through numerical examples.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 133054"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386720","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}

引用次数: 0

DyTimeNet: Dynamic cross-variable dependency network with sparse strategy for multivariate time series forecasting DyTimeNet：基于稀疏策略的多变量时间序列预测的动态交叉变量依赖网络

IF 6.5 2区计算机科学

Neurocomputing Pub Date : 2026-05-01 Epub Date: 2026-02-10 DOI: 10.1016/j.neucom.2026.132964

Ting Chen , Jinzhou Lai , Yining Sun , Wai Kin (Victor) Chan

{"title":"DyTimeNet: Dynamic cross-variable dependency network with sparse strategy for multivariate time series forecasting","authors":"Ting Chen , Jinzhou Lai , Yining Sun , Wai Kin (Victor) Chan","doi":"10.1016/j.neucom.2026.132964","DOIUrl":"10.1016/j.neucom.2026.132964","url":null,"abstract":"<div><div>Most current research on multivariate time series focuses on extracting temporal patterns, often treating the data as univariate and neglecting interactions between channels. This work introduces variable-wise auxiliary information to facilitate forecasting. In reality, channel relationships are dynamic and sparse, with changing correlations over time, and fusing all channels may distort the underlying data structure. To address this, we propose <strong>DyTimeNet</strong>, a novel dual-path learning framework designed to capture dynamic sparse correlations between variables (CrossVBlock) and complex nonlinear relationships within time series (IntraVBlock). Specifically, we introduce an innovative DropFusion sparse strategy within CrossVBlock, which simulates multiple variable-wise correlations by dynamically selecting connections (edges) and dropping neurons. By doing so, it reduces unnecessary interference from irrelevant variables, forcing the network to avoid over-relying on specific edges or neurons, thereby adapting to varying patterns of channel correlations. Experiments on thirteen real-world datasets demonstrate that DyTimeNet significantly outperforms sophisticated Transformer-based models and even surpasses LLM-based models, especially in capturing better fine-grained pattern fitting in multivariate time series. Additionally, DyTimeNet uses only 1/4 of the parameter size and trains 7<span><math><mo>×</mo></math></span> faster than state-of-the-art models like iTransformer, while improving MSE by 10.13%. These improvements make DyTimeNet ideal for applications with limited computational resources or high-frequency demands, such as edge-side sensor networks.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 132964"},"PeriodicalIF":6.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386734","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}

引用次数: 0