检索结果-内蒙古大学图书馆

A neural kernel framework for relation extraction

Expert Systems with Applications 2025年 286卷

作者： Wang, Kai Chen, Yanping Yang, Weizhe Huang, Ruizhang Qin, Yongbin Text Computing & Cognitive Intelligence Engineering Research Center of National Education Ministry College of Computer Science and Technology Guizhou University China State Key Laboratory of Public Big Data Guizhou University China

Relation extraction (RE) poses significant challenges due to the complexity of identifying semantic relationships between overlapping entity pairs within sentences. Traditional kernel-based methods effectively leverage instance-level similarities but heavily depend on manually designed kernel functions, requiring substantial domain expertise and limiting their flexibility. On the other hand, neural network approaches excel at automatically learning abstract feature representations but typically neglect critical instance-specific information, potentially missing valuable relational details during inference. These limitations motivate the exploration of a hybrid approach that effectively integrates the strengths of traditional kernels with the representational power of neural networks. In this paper, we propose a neural kernel framework designed to bridge this critical gap. The proposed framework employs neural kernels, whose parameters are initialized based on task-specific objectives and optimized through neural training procedures, enabling adaptive learning of similarity measures without relying on handcrafted kernel functions. By maintaining instance-level details through annotated training examples, neural kernels create discriminative, flexible decision boundaries tailored specifically to the relation extraction task. We further develop three complementary neural kernel components, instance kernel, description kernel, and cluster kernel, to show the advantages of kernel substitution. Extensive experiments on the ACE 2005, SemEval 2010, and CoNLL 2004 datasets demonstrate that our neural kernel framework significantly outperforms existing state-of-the-art methods, achieving F1-scores of 88.11 %, 91.08 %, and 98.63 %, respectively. © 2025 Elsevier Ltd

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

SAV-SE: Scene-aware Audio-Visual Speech Enhancement with Selective State Space Model

引用

IEEE Journal on Selected Topics in Signal Processing 2025年

作者： Qian, Xinyuan Gao, Jiaran Zhang, Yaodan Zhang, Qiquan Liu, Hexin Garcia, Leibny Paola Li, Haizhou University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China University of New South Wales School of Electrical Engineering and Telecommunications Sydney2052 Australia Nanyang Technological University College of Computing and Data Science Singapore Singapore Johns Hopkins University CLSP and HLT-COE United States Chinese University of HongKong Guangdong Provincial Key Laboratory of Big Data Computing Shenzhen518172 China Shenzhen Research Institute of Big data Shenzhen51872 China

Speech enhancement plays an essential role in various applications, and the integration of visual information has been demonstrated to bring substantial advantages. However, the majority of current research concentrates on the examination of facial and lip movements, which can be compromised or entirely inaccessible in scenarios where occlusions occur or when the camera view is distant. Whereas contextual visual cues from the surrounding environment have been overlooked: for example, when we see a dog bark, our brain has the innate ability to discern and filter out the barking noise. To this end, in this paper, we introduce a novel task, i.e. Scene-aware Audio-Visual Speech Enhancement SAV-SE. To our best knowledge, this is the first proposal to use rich contextual information from synchronized video as auxiliary cues to indicate the type of noise, which eventually improves the speech enhancement performance. Specifically, we propose the VC-S $^{2}$ E method, which incorporates the Conformer and Mamba modules for their complementary strengths. Extensive experiments are conducted on public MUSIC, AVSpeech and AudioSet datasets, where the results demonstrate the superiority of VC-S $^{2}$ E over other competitive methods. We will make the source code publicly available. Project demo page: https://***/ © 2007-2012 IEEE.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

SwimVG: Step-wise Multimodal Fusion and Adaption for Visual Grounding

arXiv

引用

arXiv 2025年

作者： Shi, Liangtao Liu, Ting Hu, Xiantao Hu, Yue Yin, Quanjun Hong, Richang Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Hefei230009 China Ministry of Education and School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China School of systems engineering National University of Defense Technology Hunan Province Changsha410073 China Department of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210014 China

Visual grounding aims to ground an image region through natural language, which heavily relies on cross-modal alignment. Most existing methods transfer visual/linguistic knowledge separately by fully fine-tuning uni-modal pre-trained models, followed by a simple stack of visual-language transformers for multimodal fusion. However, these approaches not only limit adequate interaction between visual and linguistic contexts, but also incur significant computational costs. Therefore, to address these issues, we explore a step-wise multimodal fusion and adaption framework, namely SwimVG. Specifically, SwimVG proposes step-wise multimodal prompts (Swip) and cross-modal interactive adapters (CIA) for visual grounding, replacing the cumbersome transformer stacks for multimodal fusion. Swip can improve the alignment between the vision and language representations step by step, in a token-level fusion manner. In addition, weight-level CIA further promotes multimodal fusion by cross-modal interaction. Swip and CIA are both parameter-efficient paradigms, and they fuse the cross-modal features from shallow to deep layers gradually. Experimental results on four widely-used benchmarks demonstrate that SwimVG achieves remarkable abilities and considerable benefits in terms of efficiency. Our code is available at https://***/liuting20/SwimVG. Copyright © 2025, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

multiPI-TransBTS: A multi-path learning framework for brain tumor image segmentation based on multi-physical information

引用

Computers in Biology and Medicine 2025年 191卷 110148-110148页

作者： Zhu, Hongjun Huang, Jiaohang Chen, Kuo Ying, Xuehui Qian, Ying School of Software Engineering Chongqing University of Posts and Telecommunications Chongqing400065 China Chongqing Engineering Research Center of Software Quality Assurance Testing and Assessment Chongqing400065 China Key Laboratory of Big Data Intelligent Computing Chongqing University of Posts and Telecommunications Chongqing400065 China

Brain Tumor Segmentation (BraTS) plays a critical role in clinical diagnosis, treatment planning, and monitoring the progression of brain tumors. However, due to the variability in tumor appearance, size, and intensity across different MRI modalities, automated segmentation remains a challenging task. In this study, we propose a novel Transformer-based framework, multiPI-TransBTS, which integrates multi-physical information to enhance segmentation accuracy. The model leverages spatial information, semantic information, and multi-modal imaging data, addressing the inherent heterogeneity in brain tumor characteristics. The multiPI-TransBTS framework consists of an encoder, an Adaptive Feature Fusion (AFF) module, and a multi-source, multi-scale feature decoder. The encoder incorporates a multi-branch architecture to separately extract modality-specific features from different MRI sequences. The AFF module fuses information from multiple sources using channel-wise and element-wise attention, ensuring effective feature recalibration. The decoder combines both common and task-specific features through a Task-Specific Feature Introduction (TSFI) strategy, producing accurate segmentation outputs for Whole Tumor (WT), Tumor Core (TC), and Enhancing Tumor (ET) regions. Comprehensive evaluations on the BraTS2019 and BraTS2020 datasets demonstrate the superiority of multiPI-TransBTS over the state-of-the-art methods. The model consistently achieves better Dice coefficients, Hausdorff distances, and Sensitivity scores, highlighting its effectiveness in addressing the BraTS challenges. Our results also indicate the need for further exploration of the balance between precision and recall in the ET segmentation task. The proposed framework represents a significant advancement in BraTS, with potential implications for improving clinical outcomes for brain tumor patients. © 2025

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

A Clustering-Based Color Reordering Method for Reversible data Hiding in Palette Images

引用

IET Image Processing 2025年第1期19卷

作者： Deng, Jianxuan Chen, Yi Wang, Hongxia Guo, Chun Cui, Yunhe Shen, Guowei State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guiyang China Engineering Research Center of Text Computing & Cognitive Intelligence Ministry of Education Guizhou University Guiyang China School of Cyber Science and Engineering Sichuan University Chengdu China

A recent research work pointed out that the reversible data hiding algorithms proposed for gray-scale images can be implemented on the reconstructed palette images to improve embedding capacity and visual quality by reordering the color table. However, the reordering effect has a significant impact on performance improvement. Therefore, we propose a clustering-based color reordering method for reversible data hiding in palette images to improve the reordering effect and further enhance the performance. In this method, we first design a centroid initialization method to select the initial centroids and then exploit the K-means algorithm to generate (Formula presented.) clusters for the colors in the original color table. In the following, our proposed method, respectively, reorders the colors of these clusters by a greedy strategy and concatenates them into the reordered color table. Based on the relationship between the original and the reordered color tables, a novel index matrix can be reconstructed. Finally, state-of-the-art reversible data hiding algorithms can be implemented on the reconstructed index matrix for performance improvement. Since our proposed method improves the reordering effect, enhances the correlation of the reconstructed index matrix, and reduces the length of the encoded location map, the maximal embedding capacities and the visual quality under the fixed embedding capacities are improved. We conducted experiments on two image datasets and six standard images to verify that the performance improvement of our proposed reordering method is better than that of the state-of-the-art methods. © 2025 The Author(s). IET Image Processing published by John Wiley & Sons Ltd on behalf of The Institution of engineering and Technology.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Gradient neural network models for approximate Tucker decomposition of time-dependent tensors

引用

Neurocomputing 2025年 640卷

作者： Che, Maolin Wei, Yimin Yan, Hong State Key Laboratory of Public Big Data and School of Mathematics and Statistics Guizhou University Guiyang550025 China Center for Intelligent Multidimensional Data Analysis Hong Science Park Shatin Hong Kong School of Mathematical Sciences and Key Laboratory of Mathematics for Nonlinear Sciences Fudan University Shanghai200433 China Department of Electrical Engineering and Center for Intelligent Multidimensional Data Analysis City University of Hong Kong 83 Tat Chee Avenue Kowloon Hong Kong

In this paper, we introduce a gradient neural network model for approximating the Tucker decomposition of time-dependent tensors with a specified multilinear rank bound. The behavior of this model is governed by a system of ordinary differential equations. First, based on the Karush–Kuhn–Tucker (KKT) conditions of the optimization problem for the approximate Tucker decomposition of a constant tensor, a gradient neural network model is proposed, demonstrating local stability according to Lyapunov stability theory. Next, we design a gradient neural network model specifically tailored for time-dependent tensors. Finally, using the Taylor-type difference rules, we discretize the proposed neural network models, ensuring zero stability for the implementation of digital hardware. We illustrate the effectiveness and characteristics of our approach through numerical examples. © 2025 Elsevier B.V.

关键词： Lyapunov methods

来源：评论

学校读者我要写书评

暂无评论

Global Feature Enhancing and Fusion Framework for Strain Gauge Status Recognition 25

Global Feature Enhancing and Fusion Framework for Strain Gau...

引用

Companion Proceedings of the ACM on Web Conference 2025

作者： Xu Zhang Peng Wang Chen Wang Zhe Xu Xiaohua Nie Wei Wang Shanghai Key Laboratory of Data Science School of Computer Science and Technology Fudan University Shang Hai China Shanghai Key Laboratory of Data Science School of Computer Science and Technology Fudan University Shanghai China National Engineering Research Center for Big Data Software Tsinghua University Beijing China School of Software Tsinghua University Beijing China National Key Laboratory of Strength and Structural Integrity Aircraft Strength Research Institute of China Xi'an Shaanxi China

ISBN: (纸本)9798400713316

Strain Gauge Status (SGS) recognition is crucial in the field of intelligent manufacturing based on the Internet of Things, as accurate identification helps timely detection of failed mechanical components, avoiding accidents. The loading and unloading sequences generated by strain gauges can be identified through time series classification (TSC) algorithms. Recently, deep learning models, e.g., convolutional neural networks (CNNs) have shown remarkable success in the TSC task, as they can extract discriminative local features from the subsequences to identify the time series. However, we observe only the local features may not be sufficient for expressing the time series, especially when the local sub-sequences between different time series are very similar, e.g., SGS data of aircraft wings in static strength experiments. Nevertheless, CNNs suffer from the limitation in extracting global features due to the nature of convolution operations. For extracting global features to more comprehensively represent the SGS time series, we propose two insights: (i) Constructing global features through feature engineering. (ii) Learning high-order relationships between local features to capture global features. To realize and utilize them, we propose a hypergraph-based global feature learning and fusion framework, which learns and fuses global features for semantic consistency to enhance the representation of SGS time series, thereby improving recognition accuracy. Our method designs are validated on industrial SGS and public UCR datasets, showing better generalization for unseen data in SGS recognition.

关键词： deep learning models

来源：评论

学校读者我要写书评

暂无评论

BadToken: Token-level Backdoor Attacks to Multi-modal Large Language Models

arXiv

引用

arXiv 2025年

作者： Yuan, Zenghui Shi, Jiawen Zhou, Pan Gong, Neil Zhenqiang Sun, Lichao Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University of Science and Technology China Duke University United States Lehigh University United States

Multi-modal large language models (MLLMs) extend large language models (LLMs) to process multi-modal information, enabling them to generate responses to image-text inputs. MLLMs have been incorporated into diverse multimodal applications, such as autonomous driving and medical diagnosis, via plug-and-play without fine-tuning. This deployment paradigm increases the vulnerability of MLLMs to backdoor attacks. However, existing backdoor attacks against MLLMs achieve limited effectiveness and stealthiness. In this work, we propose BadToken, the first token-level backdoor attack to MLLMs. BadToken introduces two novel backdoor behaviors: Token-substitution and Token-addition, which enable flexible and stealthy attacks by making token-level modifications to the original output for backdoored inputs. We formulate a general optimization problem that considers the two backdoor behaviors to maximize the attack effectiveness. We evaluate BadToken on two open-source MLLMs and various tasks. Our results show that our attack maintains the model’s utility while achieving high attack success rates and stealthiness. We also show the real-world threats of BadToken in two scenarios, i.e., autonomous driving and medical diagnosis. Furthermore, we consider defenses including fine-tuning and input purification. Our results highlight the threat of our attack. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

UAV-Assisted Joint Mobile Edge Computing and data Collection via Matching-Enabled Deep Reinforcement Learning

引用

IEEE Internet of Things Journal 2025年

作者： Wang, Boxiong Kang, Hui Li, Jiahui Sun, Geng Sun, Zemin Jilin University College of Computer Science and Technology Changchun130012 China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Changchun130012 China Nanyang Technological University College of Computing and Data Science 639798 Singapore

Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) and data collection (DC) have been popular research issues. Different from existing works that consider MEC and DC scenarios separately, this paper investigates a multi-UAV-assisted joint MEC-DC system. Specifically, we formulate a joint optimization problem to minimize the MEC latency and maximize the collected data volume. This problem can be classified as a non-convex mixed integer programming problem that exhibits long-term optimization and dynamics. Thus, we propose a deep reinforcement learning-based approach that jointly optimizes the UAV movement, user transmit power, and user association in real time to solve the problem efficiently. Specifically, we reformulate the optimization problem into an action space-reduced Markov decision process (MDP) and optimize the user association by using a two-phase matching-based association (TMA) strategy. Subsequently, we propose a soft actor-critic (SAC)-based approach that integrates the proposed TMA strategy (SAC-TMA) to solve the formulated joint optimization problem collaboratively. Simulation results demonstrate that the proposed SAC-TMA is able to coordinate the two subsystems and can effectively reduce the system latency and improve the data collection volume compared with other benchmark algorithms. © 2025 IEEE. All rights reserved,

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Novel 3D UAV Path Planning for IoT Services Based on Interactive Cylindrical Vector Teaching–Learning Optimization Algorithm

引用

Sensors 2025年第8期25卷 2407-2407页

作者： Jiang, Xinghe Wu, Xuanyu Zhang, Zhifeng Hong, Zhaoxi Xiao, Xi Feng, Yixiong State Key Laboratory of Public Big Data Guizhou University Guiyang550025 China School of Mechanical Engineering Guizhou University Guiyang550025 China School of Mechanical Engineering Zhejiang University Hangzhou310027 China State Key Laboratory of Fluid Power and Mechatronic Systems Zhejiang University Hangzhou310027 China Ocean College Zhejiang University Hangzhou310027 China

In the 6G-IoT convergence ecosystem, UAV path planning for static environments is systematically investigated as a resource coordination problem where communication demands and terrain constraints are balanced through intelligent trajectory optimization. The innovation of this paper lies in the proposal of an interactive cylinder vector teaching–learning-based optimization (ICVTLBO) algorithm, where UAV trajectory points are represented in cylindrical coordinates, and targeted interactive strategies are proposed during the teacher and learner phases to address uncertainty challenges, such as terrain elevation fluctuations and communication link instability caused by obstacles in static environments. The ICVTLBO is compared with other classical and novel algorithms on the CEC2022 benchmark function suite, demonstrating its effectiveness and reliability in solving complex optimization problems. Subsequently, real digital elevation model (DEM) maps are utilized to establish nine diverse terrain scenarios for the simulation of 3D UAV path planning challenges, and experimental results show that the ICVTLBO outperforms other methods, providing high-quality paths for UAVs in complex environments. © 2025 by the authors.

关键词： Target drones

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：