Query：
Refining：
Year
Type
Indexed by
Colleges
Complex
Language
Clean All
Abstract ：
A large-scale model is typically trained on an extensive dataset to update its parameters and enhance its classification capabilities. However, directly using such data can raise significant privacy concerns, especially in the medical field, where datasets often contain sensitive patient information. Federated Learning (FL) offers a solution by enabling multiple parties to collaboratively train a high-performance model without sharing their raw data. Despite this, during the federated training process, attackers can still potentially extract private information from local models. To bolster privacy protections, Differential Privacy (DP) has been introduced to FL, providing stringent safeguards. However, the combination of DP and data heterogeneity can often lead to reduced model accuracy. To tackle these challenges, we introduce a sampling-memory mechanism, FedSam, which improves the accuracy of the global model while maintaining the required noise levels for differential privacy. This mechanism also mitigates the adverse effects of data heterogeneity in heterogeneous federated environments, thereby improving the global model's overall performance. Experimental evaluations on datasets demonstrate the superiority of our approach. FedSam achieves a classification accuracy of 95.03%, significantly outperforming traditional DP-FedAvg (91.74%) under the same privacy constraints, highlighting FedSam's robustness and efficiency.
Keyword ：
Data heterogeneity Differential privacy Federated learning
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Li, Hongtao , Li, Xinyu , Liu, Ximeng et al. FedSam: Enhancing federated learning accuracy with differential privacy and data heterogeneity mitigation [J]. | COMPUTER STANDARDS & INTERFACES , 2026 , 95 . |
| MLA | Li, Hongtao et al. "FedSam: Enhancing federated learning accuracy with differential privacy and data heterogeneity mitigation" . | COMPUTER STANDARDS & INTERFACES 95 (2026) . |
| APA | Li, Hongtao , Li, Xinyu , Liu, Ximeng , Wang, Bo , Wang, Jie , Tian, Youliang . FedSam: Enhancing federated learning accuracy with differential privacy and data heterogeneity mitigation . | COMPUTER STANDARDS & INTERFACES , 2026 , 95 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Recently, deep learning-based methods have made impressive progress in infrared small target detection (IRSTD). However, the weak and variable nature of small targets constrains the feature extraction and scene adaptation of existing methods, leading to low data utilization and poor robustness. To address this issue, we innovatively introduce the feedback mechanism into IRSTD and propose the dynamic feedback iterative network (DFINet). The main motivation is to guide the model training and prediction utilizing the history prediction mask (HPMK) of previous rounds. On the one hand, in the training phase, DFINet can further mine the key features of real targets by training in multiple iterations with limited data; on the other hand, in the prediction phase, DFINet can correct the wrong results through feedback iterative to improve the model robustness. Specifically, we first propose the dynamic feedback feature fusion module (DFFFM), which dynamically interacts HPMK with feature maps through a hard attention mechanism to guide feature mining and error correction. Then, for better feature extraction, the cascaded hybrid pyramid pooling module (CHPP) is devised to capture both global and local information. Finally, we propose the dynamic semantic fusion module (DSFM), which innovatively utilizes feedback information to guide the fusion of high-level and low-level features for better feature representation in different scenarios. Extensive experimental results on publicly available datasets of NUDT-SIRST, IRSTD-1k, and SIRST Aug show that DFINet outperforms several state-of-the-art methods and achieves superior detection performance. Our code will be publicly available at https://github.com/uisdu/DFINet.
Keyword ：
Error correction Feature mining Feedback iteration Infrared small target detection
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wu, Jing , Luo, Changhai , Qiu, Zhaobing et al. DFINet: Dynamic feedback iterative network for infrared small target detection [J]. | PATTERN RECOGNITION , 2026 , 169 . |
| MLA | Wu, Jing et al. "DFINet: Dynamic feedback iterative network for infrared small target detection" . | PATTERN RECOGNITION 169 (2026) . |
| APA | Wu, Jing , Luo, Changhai , Qiu, Zhaobing , Chen, Liqiong , Ni, Rixiang , Li, Yunxiang et al. DFINet: Dynamic feedback iterative network for infrared small target detection . | PATTERN RECOGNITION , 2026 , 169 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
In this paper, we propose HSENet, a hierarchical semantic-enriched network capable of generating high-quality fused images with robust global semantic consistency and excellent local detail representation. The core innovation of HSENet lies in its hierarchical enrichment of semantic information through semantic gathering, distribution, and injection. Specifically, the network begins by balancing global information exchange via multi-scale feature aggregation and redistribution while dynamically bridging fusion and segmentation tasks. Following this, a progressive semantic dense injection strategy is introduced, employing dense connections to first inject global semantics into highly consistent infrared features and then propagate the semantic-infrared hybrid features to visible features. This approach effectively enhances semantic representation while minimizing high-frequency information loss. Furthermore, HSENet includes two types of feature fusion modules, to leverage cross-modal attention for more comprehensive feature fusion and utilize semantic features as a third input to further enhance the semantic representation for image fusion. These modules achieve robust and flexible feature fusion in complex scenarios by dynamically balancing global semantic consistency and fine-grained local detail representation. Our approach excels in visual perception tasks while fully preserving the texture features from the source modalities. The comparison experiments of image fusion and semantic segmentation demonstrate the superiority of HSENet in visual quality and semantic preservation. The code is available at https://github.com/Lxyklmyt/HSENet.
Keyword ：
High-level vision task Image fusion Progressive semantic dense injection Semantic gathering and distribution
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Liu, Xinyu , Ming, Rui , Du, Songlin et al. HSENet: Hierarchical semantic-enriched network for multi-modal image fusion [J]. | PATTERN RECOGNITION , 2026 , 170 . |
| MLA | Liu, Xinyu et al. "HSENet: Hierarchical semantic-enriched network for multi-modal image fusion" . | PATTERN RECOGNITION 170 (2026) . |
| APA | Liu, Xinyu , Ming, Rui , Du, Songlin , He, Lianghua , Xiao, Guobao . HSENet: Hierarchical semantic-enriched network for multi-modal image fusion . | PATTERN RECOGNITION , 2026 , 170 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Sorting problem has become one of the most common Multi-Criteria Decision-Making (MCDM) problems in real-life scenarios. In classical multi-criteria sorting methods, the input data are required to be precisely expressed by quantitative and crisp values. However, due to the inherent uncertainty of real-life problems, crisp values may not be enough to model the decision information. In addition, while multi-criteria sorting methods often utilize group intelligence, they seldom address the challenges of Consensus Reaching Processes (CRP) as non-cooperative behaviors or human bounded rationality, in spite of they are common challenges in MCDM. In such a context, this paper develops a novel multi-criteria sorting method to model the uncertainty, manage the CRP as well as capture the human bounded rationality, and then obtain better decision solutions by considering these issues. An illustrative numerical example is presented to demonstrate the effectiveness and applicability of the proposed method. The results highlight the potential of the proposed method in addressing the multi-criteria sorting problems with uncertain and imprecise input information under bounded rationality hypothesis. Comparative analysis and sensitivity analysis are also conducted to show the advantages and robustness of the current proposal, respectively. © 2025 Elsevier B.V.
Keyword ：
Behavioral research Decision making Numerical methods Screening Sensitivity analysis Sorting Uncertainty analysis
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Pan, Xiao-Hong , He, Shi-Fan , Wang, Ying-Ming et al. An ExpTODIM based multi-criteria sorting method under uncertainty [J]. | Information Fusion , 2026 , 125 . |
| MLA | Pan, Xiao-Hong et al. "An ExpTODIM based multi-criteria sorting method under uncertainty" . | Information Fusion 125 (2026) . |
| APA | Pan, Xiao-Hong , He, Shi-Fan , Wang, Ying-Ming , Martínez, Luis . An ExpTODIM based multi-criteria sorting method under uncertainty . | Information Fusion , 2026 , 125 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Single Image Reflection Removal (SIRR) is an active topic in low-level vision, aiming to eliminate the influence of reflected objects or light sources on image quality. However, due to the ill-posed property of SIRR and the lack of large-scale real world reflection image datasets, existing methods degrade on real datasets and suffer from the problem of reflection residue. To address these issues, we propose an effective SIRR network called PA-NAFNet. It utilizes a non-linear activation-free network (NAFNet) as the baseline and incorporates a pyramid attention module to capture long-range pixel interactions. Additionally, during the training phase, color jittering technique is introduced to increase the diversity of the training dataset, thereby alleviating potential color distortion issues after reflection removal. Experimental results on multiple reflection removal benchmark tests demonstrate the effectiveness of PA-NAFNet. The relevant code is available on this link. © 2025 Elsevier Inc.
Keyword ：
Benchmarking Chemical activation Color Color image processing Deep learning Image denoising Image enhancement Image quality Light reflection Light sources Sodium compounds
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Zhang, Qing , Zhang, Yizhong , Kuang, Xu et al. PA-NAFNet: An improved nonlinear activation free network with pyramid attention for single image reflection removal [J]. | Digital Signal Processing: A Review Journal , 2026 , 168 . |
| MLA | Zhang, Qing et al. "PA-NAFNet: An improved nonlinear activation free network with pyramid attention for single image reflection removal" . | Digital Signal Processing: A Review Journal 168 (2026) . |
| APA | Zhang, Qing , Zhang, Yizhong , Kuang, Xu , Zhou, Yuanbo , Tong, Tong . PA-NAFNet: An improved nonlinear activation free network with pyramid attention for single image reflection removal . | Digital Signal Processing: A Review Journal , 2026 , 168 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Speech disorders can significant impact speakers’ articulation, resulting in large variations in speech. These variations can affect the performance of Automatic Speech Recognition (ASR), limiting the access of individuals with speech disorders to the benefits provided by this technology. Previous research on human speech perception has shown that both auditory and articulatory information play important roles, with the latter being more effective when the input speech is distorted. When a sound is perceived, the brain processes its auditory features and activates neural simulations of the articulatory movements associated with that sound. Throughout this process, acoustic and articulatory information often enhance each other, improving the overall comprehension and processing of the auditory stimulus. Motivated by these findings, this study proposes an Inclusive Gestural Feature Extraction (InGesFE) method and a Multi-Feature Co-Activation Module (MF-CoAct) to address the challenge of large variability in dysarthric ASR. The InGesFE method extracts features using a richness constraint and a phoneme distinctiveness constraint, enabling them to share similar characteristics with articulatory gestures, including: (1) rich aspects of input speech, (2) phonemic distinctiveness, and (3) robustness in conveying intent. Meanwhile, the MF-CoAct facilitates the co-activation of auditory and articulatory (gestural) features through a statistical variable-based activation network. Additionally, a continual pre-training method is designed to support faster and more effective adaptation to highly variable speech. To evaluate the effectiveness of the proposed method, two widely used dysarthria datasets, TORGO and UASpeech, are employed. Across both datasets, our approach led to a relative word error rate reduction (WERR) of 13.75%–15.37% for single-word recognition and 36.48% for multiword recognition compared to the baseline. It outperformed existing methods for speakers with severe dysarthria and very low intelligibility, reaching a word error rate (WER) of 51.41% on the UASpeech dataset. It also demonstrated increased robustness in noisy environments, achieving a 19.16% WERR in single-word recognition and a 38.49% WERR in multiword recognition under noisy conditions. Further analysis indicates that the features extracted by InGesFE capture richer articulatory information beyond auditory features alone, particularly improving the representation of co-articulatory cues. © 2025 Elsevier B.V.
Keyword ：
Audition Chemical activation Errors Extraction Feature extraction Speech communication Speech intelligibility Speech processing Speech recognition
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Lin, Yuqin , Wang, Longbiao , Dang, Jianwu et al. Gestural feature extraction and multi-feature co-activation for dysarthric speech recognition [J]. | Information Fusion , 2026 , 125 . |
| MLA | Lin, Yuqin et al. "Gestural feature extraction and multi-feature co-activation for dysarthric speech recognition" . | Information Fusion 125 (2026) . |
| APA | Lin, Yuqin , Wang, Longbiao , Dang, Jianwu , Minematsu, Nobuaki . Gestural feature extraction and multi-feature co-activation for dysarthric speech recognition . | Information Fusion , 2026 , 125 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
The multi-turn response selection is an important component in retrieval-based human–computer dialogue systems. Most recent models adopt the utilization of pre-trained language models to acquire fine-grained semantic information within diverse dialogue contexts, thereby enhancing the precision of response selection. However, effectively leveraging the language style information of speakers along with the topic information in the dialogue context to enhance the semantic understanding capability of pre-trained language models still poses a significant challenge that requires resolution. To address this challenge, we propose a BERT-based Language Style and Topic Aware (BERT-LSTA) model for multi-turn response selection. BERT-LSTA augments BERT with two distinctive modules: the Language Style Aware (LSA) module and the Question-oriented Topic Window Selection (QTWS) module. The LSA module introduces a contrastive learning method to learn the latent language style information from distinct speakers in the dialogue. The QTWS module proposes a topic window segmentation algorithm to segment the dialogue context into topic windows, which facilitates the capacity of BERT-LSTA to refine and incorporate relevant topic information for response selection. Experimental results on two public benchmark datasets demonstrate that BERT-LSTA outperforms all state-of-the-art baseline models across various metrics. Furthermore, ablation studies reveal that the LSA module significantly improves performance by capturing speaker-specific language styles, while the QTWS module enhances topic relevance by filtering irrelevant contextual information. © 2025
Keyword ：
BERT Contrastive learning Language style Multi-turn response selection Topic window segmentation
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Li, W. , Chen, Y. , Xu, J. et al. Multi-turn response selection with Language Style and Topic Aware enhancement [J]. | Computer Speech and Language , 2026 , 95 . |
| MLA | Li, W. et al. "Multi-turn response selection with Language Style and Topic Aware enhancement" . | Computer Speech and Language 95 (2026) . |
| APA | Li, W. , Chen, Y. , Xu, J. , Zhong, J. , Dong, C. . Multi-turn response selection with Language Style and Topic Aware enhancement . | Computer Speech and Language , 2026 , 95 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Optical instruments, electronic products, and medical fields urgently require the development of robust and transparent anti-fouling coatings. However, traditional coatings often struggle to balance mechanical properties with anti-fouling performance and may contain components that are detrimental to the environment. A high cross-linking density design helps to enhance the mechanical properties and corrosion resistance of coatings without the use of environmentally unfriendly components. In this work, we prepared a novel polyhedral oligomeric silsesquioxane (POSS) material with multiple epoxy groups, EP-POSS, through a click reaction. We selected an appropriate multi-point amine curing agent, tetraethylenepentamine, leveraging its efficient curing with EP-POSS to construct a multi-point, highly cross-linked network. This process not only strengthens the mechanical strength and chemical resistance of the coating but is also simple and cost-effective. Environmentally friendly polydimethylsiloxane is used to provide flexibility and liquid repellency. The final EPOSSPT coating boasts high transparency (> 99 %), low surface roughness (arithmetic mean deviation (Sa): ∼0.515 nm), and excellent repulsion to both liquid and solid contaminants, demonstrating superior self-cleaning and anti-fouling properties. It achieves a balance between high hardness (7H) and excellent flexibility (7500 bending cycles with a bending radius of 1 mm), while also resisting wear and impact from everyday use. This high-transparency, robust, and flexible coating has broad application prospects and offers insights into the design of new high-performance protective materials. © 2025
Keyword ：
Flexible liquid-like coating Fluoride-free High hardness Highly cross-linked networks Highly transparent
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Xie, Y. , Wu, S. , Liu, H. et al. High-performance anti-fouling coating: Achieving high transparency, durability, and flexibility via epoxy-amine curing with dense cross-linking design [J]. | Journal of Materials Science and Technology , 2026 , 242 : 255-263 . |
| MLA | Xie, Y. et al. "High-performance anti-fouling coating: Achieving high transparency, durability, and flexibility via epoxy-amine curing with dense cross-linking design" . | Journal of Materials Science and Technology 242 (2026) : 255-263 . |
| APA | Xie, Y. , Wu, S. , Liu, H. , Zheng, Y. , Zhao, K. , Huang, J. et al. High-performance anti-fouling coating: Achieving high transparency, durability, and flexibility via epoxy-amine curing with dense cross-linking design . | Journal of Materials Science and Technology , 2026 , 242 , 255-263 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Aqueous zinc-ion batteries (AZIBs) hold promising prospects for large-scale energy storage systems, yet their commercialization is hindered by dendritic growth and water-induced side reactions associated with zinc anodes, especially at high depths of discharge (DOD). Herein, a multifunctional zincophilic additive is developed to promote the planar Zn deposition and construct a stable solid electrolyte interphase (SEI). Disodium malate (DMA) possesses pH-buffering capability that maintains electrolyte pH stability during prolonged cycling, effectively mitigating side reactions. Furthermore, the concentration of DMA significantly influences crystal deposition. An appropriate amount of DMA molecules selectively adsorbs onto the zinc foil, facilitating uniform zinc ion deposition on the (002) crystal plane. In addition, disodium maleate molecules reconfigure the electric double layer (EDL) to reduce free water interaction and promote the in-situ formation of the dense SEI, consisting of inorganic zinc salt and amorphous organic component, on the Zn metal surface. Notably, the dense organic-inorganic hybrid SEI layer persists with remarkable structural integrity even after long cycling. These features enable a highly reversible dendrite-free Zn plating/stripping process and suppress side reactions. As a result, Zn||Zn cells with DMA additives demonstrate extended cycling stability, enduring up to 5000 h at 8.6% DOD. Moreover, DMA-modified Zn anodes achieve an exceptional cycle lifespan of 750 h under 81.9% DOD with a high coulombic efficiency of 99.81% in asymmetric cells. In full-cell configurations, Zn||I2 cells stably cycle for over 12,000 cycles, retaining 89.77% of their capacity. This electrolyte regulation strategy offers a compelling pathway for the development of aqueous zinc ion batteries. © 2025
Keyword ：
Zinc powder metallurgy
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Ji, Haojie , Liang, Yuhang , Yang, Tao et al. Dense solid electrolyte interphase and Zn (002) plane texture enabling high depth-of-discharge anode for highly reversible zinc ion batteries [J]. | Journal of Materials Science and Technology , 2026 , 240 : 56-64 . |
| MLA | Ji, Haojie et al. "Dense solid electrolyte interphase and Zn (002) plane texture enabling high depth-of-discharge anode for highly reversible zinc ion batteries" . | Journal of Materials Science and Technology 240 (2026) : 56-64 . |
| APA | Ji, Haojie , Liang, Yuhang , Yang, Tao , Wu, Hongbo , Sheng, Ouwei , Shen, Tianyu et al. Dense solid electrolyte interphase and Zn (002) plane texture enabling high depth-of-discharge anode for highly reversible zinc ion batteries . | Journal of Materials Science and Technology , 2026 , 240 , 56-64 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Aqueous zinc-ion batteries are highly favored for grid-level energy storage owing to their low cost and high safety,but their practical application is limited by slow ion migration.To address this,a strategy has been developed to create a cation-accelerating electric field on the surface of the cathode to achieve ultrafast Zn2+diffusion kinetics.By employing electrodeposition to coat MoS2 on the surface of BaV6O16·3H2O nanowires,the directional built-in electric field generated at the heterointerface acts as a cation accelerator,continuously accelerating Zn2+diffusion into the active material.The optimized Zn2+diffusion coefficient in CC@BaV6O16-3H2O@MoS2(7.5 × 10-8 cm2 s-1)surpasses that of most reported V-based cathodes.Simultaneously,MoS2 serving as a cathodic armor extends the cycling life of the Zn-CC@BaV6O16·3H2O@MoS2 full batteries to over 10000 cycles.This work provides valuable insights into optimizing ion diffusion kinetics for high-performance energy storage devices.
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Yawei Xiao , Qianqian Gu , Haoyu Li et al. Design of a cationic accelerator enabling ultrafast ion diffusion kinetics in aqueous zinc-ion batteries [J]. | 能源化学 , 2025 , 100 (1) : 377-384 . |
| MLA | Yawei Xiao et al. "Design of a cationic accelerator enabling ultrafast ion diffusion kinetics in aqueous zinc-ion batteries" . | 能源化学 100 . 1 (2025) : 377-384 . |
| APA | Yawei Xiao , Qianqian Gu , Haoyu Li , Mengyao Li , Yude Wang . Design of a cationic accelerator enabling ultrafast ion diffusion kinetics in aqueous zinc-ion batteries . | 能源化学 , 2025 , 100 (1) , 377-384 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
