检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Zhang, Yongheng Yan, Danfeng State Key Laboratory of Networking and Switching Technology BUPT Beijing China

Model compression through knowledge distillation has seen extensive application in classification and segmentation tasks. However, its potential in image-to-image translation, particularly in image restoration, remains underexplored. To address this gap, we propose a Simultaneous Learning Knowledge Distillation (SLKD) framework tailored for model compression in image restoration tasks. SLKD employs a dual-teacher, single-student architecture with two distinct learning strategies: Degradation Removal Learning (DRL) and Image Reconstruction Learning (IRL), simultaneously. In DRL, the student encoder learns from Teacher A to focus on removing degradation factors, guided by a novel BRISQUE extractor. In IRL, the student decoder learns from Teacher B to reconstruct clean images, with the assistance of a proposed PIQE extractor. These strategies enable the student to learn from degraded and clean images simultaneously, ensuring high-quality compression of image restoration models. Experimental results across five datasets and three tasks demonstrate that SLKD achieves substantial reductions in FLOPs and parameters, exceeding 80%, while maintaining strong image restoration performance. Copyright © 2025, The Authors. All rights reserved.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Soft Knowledge Distillation with Multi-Dimensional Cross-Net Attention for Image Restoration Models Compression

Soft Knowledge Distillation with Multi-Dimensional Cross-Net...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yongheng Zhang Danfeng Yan State Key Laboratory of Networking and Switching Technology BUPT Beijing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Transformer-based encoder-decoder models have achieved remarkable success in image-to-image transfer tasks, particularly in image restoration. However, their high computational complexity—manifested in elevated FLOPs and parameter counts—limits their application in real-world scenarios. Existing knowledge distillation methods in image restoration typically employ lightweight student models that directly mimic the intermediate features and reconstruction results of the teacher, overlooking the implicit attention relationships between them. To address this, we propose a Soft Knowledge Distillation (SKD) strategy that incorporates a Multi-dimensional Cross-net Attention (MCA) mechanism for compressing image restoration models. This mechanism facilitates interaction between the student and teacher across both channel and spatial dimensions, enabling the student to implicitly learn the attention matrices. Additionally, we employ a Gaussian kernel function to measure the distance between student and teacher features in kernel space, ensuring stable and efficient feature learning. To further enhance the quality of reconstructed images, we replace the commonly used L1 or KL divergence loss with a contrastive learning loss at the image level. Experiments on three tasks—image deraining, deblurring, and denoising—demonstrate that our SKD strategy significantly reduces computational complexity while maintaining strong image restoration capabilities.

关键词： Representation learning Image coding Computational modeling Contrastive learning Signal processing Transformers Kernel Computational complexity Speech processing Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Propagation dynamics of vacillating individual with behavioral variation on double-layer social network

引用

PHYSICS LETTERS A 2025年 530卷

作者： Xu, Wenlong Lu, Kejian Zhu, Xuzhen Beijing Univ Post & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

This paper investigates behavior propagation in a two-layer network, emphasizing the roles of behavioral vacillation and dependency psychology. By integrating a threshold function model and percolation theory, the study explores how varying degrees of vacillation and network viscosity impact behavior spread. The results show that higher behavioral vacillation accelerates propagation, reduces uncertainty, and lowers the critical threshold, highlighting the significant influence of psychological factors on diffusion. These findings provide valuable insights for real-world applications in marketing strategies, social influence, and public opinion dynamics. Future work should incorporate dynamic network elements to offer a more comprehensive understanding of behavior propagation.

关键词： Vacillation Double-layer network Network viscosity Dependency psychology

来源：评论

学校读者我要写书评

暂无评论

Channel Code-Book (CCB): Semantic Image-Adaptive Transmission in Satellite-Ground Scenario

引用

SENSORS 2025年第1期25卷 269-269页

作者： Cao, Hui Han, Shujun Meng, Rui Xu, Xiaodong Zhang, Ping Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

Satellite-ground communication is a critical component in the global communication system, significantly contributing to environmental monitoring, radio and television broadcasting, aerospace operations, and other domains. However, the technology encounters challenges in data transmission efficiency, due to the drastic alterations in the communication channel caused by the rapid movement of satellites. In comparison to traditional transmission methods, semantic communication (SemCom) technology enhances transmission efficiency by comprehending and leveraging the intrinsic meaning of information, making it ideal for image transmission in satellite communications. Nevertheless, current SemCom methods still struggle to adapt to varying channel conditions. To address this, we propose a SemCom transmission model based on a Channel Code-Book (CCB) for adaptive image transmission in diverse channel environments. Our model reconstructs and restores the original image by documenting fading and noise states under various channel conditions and dynamically adjusting the denoiser's model parameters. Extensive experimental results demonstrate that our CCB model outperforms three representative baseline models, including Deep JSCC, ASCN, and WITT in various environments and task conditions, achieving an advantage of more than 10 dB under high signal-to-noise ratio conditions.

关键词： semantic communication satellite-ground scenarios channel adaptive image transmission

来源：评论

学校读者我要写书评

暂无评论

TQAgent: Enhancing Table-Based Question Answering with Knowledge Graphs and Tree-Structured Reasoning

引用

APPLIED SCIENCES-BASEL 2025年第7期15卷 3788-3788页

作者： Zhao, Jianbin Zhang, Pengfei Wang, Yuzhen Xin, Rui Lu, Xiuyuan Li, Ripeng Lyu, Shuai Ou, Zhonghong Song, Meina State Grid Hebei Informat & Telecommun Branch Shijiazhuang 050051 Peoples R China Beijing Univ Posts & Telecommun Sch Comp Sci Natl Pilot Software Engn Sch Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

Table-based question answering (TableQA) has emerged as an important task in natural language processing, yet existing models face challenges in handling complex reasoning and mitigating hallucinations, especially when dealing with diverse table structures. We introduce TQAgent, a framework designed to enhance table-based reasoning by incorporating knowledge graphs and tree-structured reasoning paths. TQAgent reduces hallucinations and improves model reliability by grounding reasoning in external knowledge and dynamically sampling high-confidence paths. Additionally, it employs knowledge distillation techniques for lightweight deployment. Experimental results on the TabFact, WikiTQ, and FeTaQA datasets show significant performance improvements, with accuracy increases of up to 4% over baseline models. TQAgent's dynamic operation planning and knowledge graph integration enable effective multi-step reasoning and better handling of diverse table data. Furthermore, the framework achieves state-of-the-art results, surpassing traditional large-scale models in both reasoning accuracy and computational efficiency. These findings open new avenues for future research in table-based question answering and model deployment optimization.

关键词： table-based question answering knowledge graphs multi-step reasoning knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

Towards Bare-Hand Interaction for Whiteboard Collaboration in Virtual Reality

引用

Proceedings of the ACM on Human-Computer Interaction 2025年第2期9卷 1-31页

作者： Liu, Guangtian Su, Haonan Wang, Jingyu Qi, Qi Sun, Haifeng Zhuang, Zirui Ren, Pengfei Liao, Jianxin State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Haidian Qu Beijing China Pengcheng Laboratory Shenzhen China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China *** Beijing China

Whiteboard collaboration in virtual reality (VR) is an important task in collaborative virtual environments. The current research mainly relies on the use of controllers or dedicated pens but additional devices will cause inconvenience to users. Bare-hand writing offers rich collaborative semantics through natural gestures but remains underexplored. This paper addresses challenges and solutions for bare-hand whiteboard collaboration. We analyze the input process and identify key challenges in determining pen-drop, writing, and pen-lift intentions while maintaining user control over their avatar. Our approach addresses two VR scenarios: one without and one with physical planes. The method for the first case is called Air-writing, which dynamically adjusts the distance between the avatar’s torso and the virtual whiteboard during the processes of pen-drop and pen-lift to ensure a consistent writing experience in VR. The method for the second case is called Physical-writing, which allows users to write smoothly with passive haptic feedback and physical constraints provided by the real surface by remapping the whiteboard in VR with a plane in reality. A comprehensive user study is conducted to evaluate communication efficiency, input accuracy, collaboration efficiency, and user experience of the two methods. The experimental results indicate that bare-hand interaction improves communication efficiency by 8% over controllers and performs similarly to real-world whiteboard collaboration. The Physical-writing method also demonstrates higher accuracy and user satisfaction compared to the Air-writing method. © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

Resource allocation and cost optimization in relay-assisted mobile edge computing

引用

COMPUTING 2025年第5期107卷 1-25页

作者： Zhan, Huifang Wu, Guilu Li, Zhengquan Nie, Gaofeng Jiangnan Univ Sch Internet Things Engn Wuxi Peoples R China Beijing Univ Posts & Telecommun State key Lab Networking & Switching Technol Beijing Peoples R China Hainan Univ Sch Informat & Commun Engn Hainan Peoples R China

Motivated by the limited computing and storage resources of smart devices (SDs), existing resource management schemes under the traditional mobile edge computing (MEC) system couldn't fully utilize the network resources. This article proposes a novel relay-assisted MEC architecture to provision SDs, where an idle relay device assumes the dual role of providing fee-based communication and computing services. At the same time, tasks generated by applications exhibit three distinct processing modes: local execution, relay offloading, and edge computing. To minimize the execution cost associated with communication and computing services from the user's perspective, joint computation offloading, relay selection strategy, computation resource allocation, and spectrum resource allocation (JORA) problem is formulated, which is a non-convex problem. Subsequently, a two-stage iterative algorithm is designed to solve the JORA problem using block coordinate descent, reconstruction linearization technology, and convex optimization methods. Simulation results show the effectiveness of the proposed JORA algorithm, which can significantly save the execution cost for edge users.

关键词： Mobile edge computing Relay-assisted computing offloading Execution cost Resource allocation Block coordinate descent

来源：评论

学校读者我要写书评

暂无评论

Sea-Surface Small Target Detection Using Spiking Neural Network With Controllable False Alarm

引用

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 2025年 22卷

作者： Jiao, Yang Wang, Zeyu Wang, Dewu Xu, Shuwen Beijing Univ Posts & Telecommun Sch Informat & Commun Engn State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China Beijing Inst Technol Beijing 100081 Peoples R China Xidian Univ Natl Lab Radar Signal Proc Xian 710071 Shaanxi Peoples R China

Convolutional neural network (CNN)-based detectors for small targets on the sea surface have proven effective, yet their increasingly complex structures and high energy demands pose challenges to deployment on resource-limited devices. To address this issue, this letter proposes a residual spiking network (RSN), which is developed within the advanced framework of spiking neural networks (SNNs). By using spiking neurons (SNs) as core computational units, the RSN can efficiently transmit features extracted from time-frequency graphs (TFGs) through discrete spikes in the backbone network. Meanwhile, the output is derived from the membrane voltage of the SNs, enabling reliable control over false alarms. Experimental results from the IPIX dataset validate the RSN's stability in controlling the probability of false alarms (PFAs). With an observation time of 1.024 s and a PFA of 0.001, the average probability of detection (PD) is 0.8266. The RSN demonstrates a PD comparable to that of ResNet18 while consuming only 70% of its energy, highlighting its potential for practical applications in resource-constrained environments.

关键词： Feature extraction Clutter Training Radar Time-frequency analysis Sea surface Object detection Neurons Vectors Spiking neural networks Controllable false alarm deep learning sea-surface small target spiking neural network (SNN) time-frequency graph (TFG)

来源：评论

学校读者我要写书评

暂无评论

Decentralized Physical Infrastructure Networks (DePIN): Challenges and Opportunities

引用

IEEE NETWORK 2025年第2期39卷 91-99页

作者： Lin, Zhibin Wang, Taotao Shi, Long Zhang, Shengli Cao, Bin Shenzhen Univ Coll Elect & Informat Engn Shenzhen 518000 Peoples R China Nanjing Univ Sci & Technol Sch Elect & Opt Engn Nanjing 210094 Peoples R China Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100088 Peoples R China

The widespread use of the Internet has posed challenges to existing centralized physical infrastructure networks. Issues such as data privacy risks, service disruptions, and substantial expansion costs have emerged. To address these challenges, an innovative network architecture called Decentralized Physical Infrastructure Networks (DePIN) has emerged. DePIN leverages blockchain technology to decentralize the control and management of physical devices, addressing limitations of traditional infrastructure network. This article provides a comprehensive exploration of DePIN, presenting its five-layer architecture, key design principles. Furthermore, it presents a detailed survey of the extant applications, operating mechanisms, and provides an in-depth analysis of market data pertaining to DePIN. Finally, it discusses a wide range of the open challenges faced by DePIN.

关键词： Blockchains Computer architecture Decision making Security Scalability Data privacy Costs Web 2.0 Servers Privacy

来源：评论

学校读者我要写书评

暂无评论

Learning-Driven Swarm Intelligence: Enabling Deterministic Flows Scheduling in LEO Satellite Networks

引用

IEEE TRANSACTIONS ON MOBILE COMPUTING 2025年第5期24卷 3962-3978页

作者： Wang, Zunliang Yao, Haipeng Mai, Tianle Li, Zhipei Chen, C. L. Philip Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China Beijing Inst Technol Sch Informat & Elect Beijing 100081 Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou 510641 Peoples R China

Over the past decade, low-Earth-orbit (LEO) satellite networks have emerged as a critical infrastructure in communication systems, providing wide coverage, high reliability, and global connectivity. Recently, the development of 6G technologies has challenged the LEO satellite networks to guarantee deterministic scheduling for time-sensitive services. However, traditional deterministic networking techniques fall short for LEO satellite networks. First, these techniques impose strict time constraints, but in LEO satellite networks, delay and jitter typically range in the tens of milliseconds, which exceed these limits and render them infeasible. Second, the dynamic topologies of LEO satellite networks challenge the inflexible scheduling strategies generated by these techniques, leading to sub-optimal performance and potential strategy failures. To tackle the first problem, we propose a Cycle Specified Queuing and Forwarding (CSQF) based deterministic flows scheduling mechanism. It relaxes strict time constraints by employing cyclic multi-queue scheduling, enabling more flexible and reliable long-distance transmission. For the second problem, we propose a learning-based swarm intelligence method for deterministic flows scheduling in dynamic LEO satellite networks. It includes an algorithm that combines a Dynamic Graph Convolutional Network (DGCN) with an Adaptive Ant Colony Optimization (ACO) algorithm, referred to as the DGCN-ACO algorithm. The DGCN captures the dynamic feature of the network and generates the heuristic information. The Adaptive ACO utilizes the heuristic information and considers each flow's attribute to generate multi-path scheduling strategies for each deterministic flow, as well as updates the DGCN. The experiment results demonstrate the effectiveness of our proposed algorithm.

关键词： Low-Earth-orbit satellite networks deterministic flows scheduling cycle specified queuing and forwarding mechanism cycle specified queuing and forwarding mechanism graph neural network graph neural network ant colony optimization ant colony optimization ant colony optimization

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：