检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Cao, Zouying Yang, Yifei Zhao, Hai Department of Computer Science and Engineering Shanghai Jiao Tong University Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University Shanghai Key Laboratory of Trusted Data Circulation and Governance in Web3 China

Safety alignment is indispensable for Large Language Models (LLMs) to defend threats from malicious instructions. However, recent researches reveal safety-aligned LLMs prone to reject benign queries due to the exaggerated safety issue, limiting their helpfulness. In this paper, we propose a Safety-Conscious Activation Steering (SCANS) method to mitigate the exaggerated safety concerns in aligned LLMs. First, SCANS extracts the refusal steering vectors within the activation space and utilizes vocabulary projection to anchor some specific safety-critical layers which influence model refusal behavior. Second, by tracking the hidden state transition, SCANS identifies the steering direction and steers the model behavior accordingly, achieving a balance between exaggerated safety and adequate safety. Experiments show that SCANS achieves new state-of-the-art performance on XSTest and OKTest benchmarks, without impairing their defense capability against harmful queries and maintaining almost unchanged model capability. Copyright © 2024, The Authors. All rights reserved.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Service Function Chain Dynamic Scheduling in Space-Air-Ground Integrated Networks

引用

IEEE Transactions on Vehicular Technology 2025年

作者： Jia, Ziye Cao, Yilu He, Lijun Wu, Qihui Zhu, Qiuming Niyato, Dusit Han, Zhu Nanjing University of Aeronautics and Astronautics College of Electronic and Information Engineering Nanjing211106 China Xidian University State Key Laboratory of ISN Xian710071 China China University of Mining and Technology School of Information and Control Engineering Xuzhou221116 China Nanyang Technological University School of Computer Science and Engineering Singapore639798 Singapore University of Houston Department of Electrical and Computer Engineering HoustonTX77004 United States Kyung Hee University Department of Computer Science and Engineering Seoul446-701 Korea Republic of

As an important component of the sixth generation communication technologies, the space-air-ground integrated network (SAGIN) attracts increasing attentions in recent years. However, due to the mobility and heterogeneity of the components such as satellites and unmanned aerial vehiclesin multi-layer SAGIN, the challenges of inefficient resource allocation and management complexity are aggregated. To this end, the network function virtualization technology is introduced and can be implemented via service function chains (SFCs) deployment. However, urgent unexpected tasks may bring conflicts and resource competition during SFC deployment, and how to schedule the SFCs of multiple tasks in SAGIN is a key issue. In this paper, we address the dynamic and complexity of SAGIN by presenting a reconfigurable time extension graph and further propose the dynamic SFC scheduling model. Then, we formulate the SFC scheduling problem to maximize the number of successful deployed SFCs within limited resources and time horizons. Since the problem is in the form of integer linear programming and intractable to solve, we propose the algorithm by incorporating deep reinforcement learning. Finally, simulation results show that the proposed algorithm has better convergence and performance compared to other benchmark algorithms. © 2025 IEEE.

关键词： Network function virtualization

来源：评论

学校读者我要写书评

暂无评论

Once Read is Enough: Domain-specific Pretraining-free Language Models with Cluster-guided Sparse Experts for Long-tail Domain Knowledge 38

Once Read is Enough: Domain-specific Pretraining-free Langua...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Dong, Fang Chen, Mengyi Zhou, Jixian Shi, Yubin Chen, Yixuan Dong, Mingzhi Wang, Yujiang Li, Dongsheng Yang, Xiaochen Zhu, Rui Dick, Robert Lv, Qin Yang, Fan Lu, Tun Gu, Ning Shang, Li China and Shanghai Key Laboratory of Data Science School of Computer Science Fudan University China Oxford Suzhou Centre for Advanced Research Suzhou China University of Bath United Kingdom Microsoft Research Asia Shanghai China School of Mathematics & Statistics University of Glasgow United Kingdom Bayes Business School City St George's University of London United Kingdom Department of Electrical Engineering and Computer Science University of Michigan United States Department of Computer Science University of Colorado Boulder United States School of Microelectronics Fudan University China

Language models (LMs) only pretrained on a general and massive corpus usually cannot attain satisfying performance on domain-specific downstream tasks, and hence, applying domain-specific pretraining to LMs is a common and indispensable practice. However, domain-specific pretraining can be costly and time-consuming, hindering LMs' deployment in real-world applications. In this work, we consider the incapability to memorize domain-specific knowledge embedded in the general corpus with rare occurrences and "long-tail" distributions as the leading cause for pretrained LMs' inferior downstream performance. Analysis of Neural Tangent Kernels (NTKs) reveals that those long-tail data are commonly overlooked in the model's gradient updates and, consequently, are not effectively memorized, leading to poor domain-specific downstream performance. Based on the intuition that data with similar semantic meaning are closer in the embedding space, we devise a Cluster-guided Sparse Expert (CSE) layer to actively learn long-tail domain knowledge typically neglected in previous pretrained LMs. During pretraining, a CSE layer efficiently clusters domain knowledge together and assigns long-tail knowledge to designate extra experts. CSE is also a lightweight structure that only needs to be incorporated in several deep layers. With our training strategy, we found that during pretraining, data of long-tail knowledge gradually formulate isolated, "outlier" clusters in an LM's representation spaces, especially in deeper layers. Our experimental results show that only pretraining CSE-based LMs is enough to achieve superior performance than regularly pretrained-finetuned LMs on various downstream tasks, implying the prospects of domain-specific-pretraining-free language models. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

M2G4RTP: A Multi-Level and Multi-Task Graph Model for Instant-Logistics Route and Time Joint Prediction

M2G4RTP: A Multi-Level and Multi-Task Graph Model for Instan...

引用

International Conference on data engineering

作者： Tianyue Cai Huaiyu Wan Fan Wu Haomin Wen Shengnan Guo Lixia Wu Haoyuan Hu Youfang Lin School of Computer and Information Technology Beijing Jiaotong University Beijing China Artificial Intelligence Department Cainiao Network Hangzhou China Beijing Key Laboratory of Traffic Data Analysis and Mining Beijing China

Instant-logistics (e.g., food delivery and package pick-up) is increasingly calling for Route and Time Prediction (RTP), which aims to predict both future route and arrival time of a courier’s unvisited locations. Accurate RTP can greatly benefit the platform, such as optimizing order dispatching and improving user experience. Although recent years have witnessed various works for solving the RTP problem, they still suffer from the following three limitations: i) Failing to consider the high-level transfer mode of couriers between AOIs (Areas Of Interest, such as residential quarters or office buildings), which can help to build more accurate RTP. ii) Failing to simultaneously make the route and time prediction. Existing works either separately predict route/time or predict them in a two-step way. However, since route and time are strongly correlated (nearby locations in the route should have similar arrival times), jointly predicting them should be more effective. iii) The widely adopted tree-based or sequence-based architecture fails to fully encode the spatial relationship between different locations. To address the above limitations, we propose a multi-level and multi-task graph model, named M 2 G4RTP, for instant-logistics route and time joint prediction. Specifically, we propose a multi-level graph encoder equipped with a newly-designed GAT-e encoding module to capture couriers’ both high-level transfer modes between AOIs and low-level transfer modes between locations. Moreover, a multi-task decoder is presented to jointly predict the route and time at different levels. Finally, a loss weighting method based on homoscedastic uncertainty is designed to balance the two tasks adaptively. Extensive experiments on an industry-scale real-world dataset, as well as the online deployment on Cainiao Alibaba, demonstrate the superiority of our proposed model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Boolean matrix compressed sensing

Boolean matrix compressed sensing

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Liu, Qiang Soleymani, Mahdi Mahdavifar, Hessam National Laboratory of Industrial Control Technology Institute of Cyber-Systems and Control Zhejiang University Hangzhou310027 China Halıcıoğlu Data Science Institute University of California La Jolla San DiegoCA92093 United States Department of Electrical and Computer Engineering Northeastern University BostonMA02115 United States

ISBN: (纸本)9798350368741

In real-world datasets, leveraging the low-rank and sparsity properties enables developing efficient algorithms across a diverse array of data-related tasks, including compression, compressed sensing, matrix completion, etc. Notably, these two properties often coexist in certain real-world datasets, especially in Boolean datasets and quantized real-valued datasets. To harness the advantages of low-rank and sparsity simultaneously, we adopt a technique inspired by compressed sensing and Boolean matrix completion. Our approach entails compressing a low-rank sparse Boolean matrix by performing inner product operations with a randomly generated Boolean matrix. We then propose a decoding algorithms based on message-passing techniques to recover the original matrix. Our experiments demonstrate superior recovery performance of our proposed algorithms compared to Boolean matrix completion, with equal measurement requirements. © 2025 IEEE.

关键词： Belief Propagation Boolean Matrix Compressed Sensing Matrix Completion

来源：评论

学校读者我要写书评

暂无评论

Multimodal Identity Recognition of Spiking Neural Network Based on Brain-Inspired Reward Propagation

Multimodal Identity Recognition of Spiking Neural Network Ba...

引用

International Symposium on Parallel and Distributed Processing with Applications, ISPA

作者： Yang Liu Hongxia Zhang Yahui Li Quanqiang Wang Cheng Liu Henan Key Laboratory of Big Data Analysis and Processing Henan University Kaifeng China School of Computer and Information Engineering Henan University Kaifeng China Department of Clinical Laboratory Huaihe Hospital of Henan University Kaifeng China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

Deep learning has strongly promoted the research progress of unimodal feature learning such as face recognition and speech recognition, but it still needs to be strengthened in multimodal data representation learning, especially the problem of high power consumption, which is not conducive to the real-time application of mobile devices under limited resources. In order to realize multimodal identity recognition for low-energy edge smart devices, this paper proposes a multimodal identity recognition method of spiking neural network based on brain-like mechanism (SNN-MIR-BiM). Simulating the attention mechanism of the cognitive system to specific information improves the response of multimodal features, mimics the regulation mechanism of the neurotransmitter dopamine in the nervous system, and releases reward signals from the recognition results, which affects the neuronal state and regulates the synaptic weights through global propagation, and then optimizes the network structure. Experimental results on large-scale audio and video datasets show that the proposed method can reach 98.94% recognition accuracy, and the energy consumption of the model is about 1/10 of that of CNN under the same architecture, which verifies the effectiveness of the attention mechanism and reward propagation in network design. This research is of great practical significance in the industry applications of unmanned autonomous systems such as human-computer interaction of intelligent machines and intelligent security.

关键词： Energy consumption Neurotransmitters Attention mechanisms Accuracy Spiking neural networks Speech recognition Brain modeling Feature extraction Regulation Biological neural networks

来源：评论

学校读者我要写书评

暂无评论

PG-YOLO: A Lightweight Vehicle Detection Method for Drones

PG-YOLO: A Lightweight Vehicle Detection Method for Drones

引用

International Symposium on Parallel and Distributed Processing with Applications, ISPA

作者： Yang Liu Tengyu Fan Kun Mei Liming Zhou Henan Key Laboratory of Big Data Analysis and Processing Henan University Kaifeng China School of Computer and Information Engineering Henan University Kaifeng China Department of Clinical Laboratory Huaihe Hospital of Henan University Kaifeng China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

With the rapid development of drone technology, object detection technology is widely applied in the field of drone intelligent transportation. In response to the problems of low accuracy and high computational complexity in vehicle detection tasks on resource limited drones, this paper proposes a lightweight and efficient improved model PG-YOLO for drone vehicle detection based on the YOLOv7-tiny model. Firstly, replace the LeakyReLU activation function with the SiLU activation function to compensate for the lack of nonlinear expression in the LeakyReLU activation function, and improve the convergence speed and generalization of the model during training; Secondly, the feature extraction and fusion modules were optimized to enhance the network’s perception and capture capabilities of the target, improve detection accuracy, and reduce the number of parameters; Then, lightweight convolution PConv and GSConv are introduced, and a lightweight module PPC based on PConv is proposed to improve the network’s Neck layer, reducing the number of parameters and computation while maintain detection performance; Finally, the CBAM attention mechanism is introduced to improve the detection performance for the problem of dense vehicle targets. The improved network achieved an average detection accuracy of 60.1%, an increase of 12.1 percentage points compared to the original network, a reduction of 73.6% in parameter count, a reduction of 15.9% in computational complexity, and a detection speed of 154 FPS. Compared to the original network, the network is more lightweight and achieves higher detection accuracy, which can more effectively perform unmanned aerial vehicle detection tasks.

关键词： Training Accuracy Attention mechanisms Vehicle detection Computational modeling Transportation Feature extraction Autonomous aerial vehicles Computational complexity Drones

来源：评论

学校读者我要写书评

暂无评论

Advancements in Content-Based Image Retrieval: A Comprehensive Survey of Relevance Feedback Techniques

arXiv

引用

arXiv 2023年

作者： Qazanfari, Hamed Alyan Nezhadi, Mohammad M. Khoshdaregi, Zohreh Nozari Image Processing and Data Mining Lab Shahrood University of Technology Shahrood Iran Faculty of Science University of Science and Technology of Mazandaran Behshahr Iran Department of Computer Engineering University of Bojnord Iran

Content-based image retrieval (CBIR) systems have emerged as crucial tools in the field of computer vision, allowing for image search based on visual content rather than relying solely on metadata. This survey paper presents a comprehensive overview of CBIR, emphasizing its role in object detection and its potential to identify and retrieve visually similar images based on content features. Challenges faced by CBIR systems, including the semantic gap and scalability, are discussed, along with potential solutions. It elaborates on the semantic gap, which arises from the disparity between low-level features and high-level semantic concepts, and explores approaches to bridge this gap. One notable solution is the integration of relevance feedback (RF), empowering users to provide feedback on retrieved images and refine search results iteratively. The survey encompasses long-term and short-term learning approaches that leverage RF for enhanced CBIR accuracy and relevance. These methods focus on weight optimization and the utilization of active learning algorithms to select samples for training classifiers. Furthermore, the paper investigates machine learning techniques and the utilization of deep learning and convolutional neural networks to enhance CBIR performance. This survey paper plays a significant role in advancing the understanding of CBIR and RF techniques. It guides researchers and practitioners in comprehending existing methodologies, challenges, and potential solutions while fostering knowledge dissemination and identifying research gaps. By addressing future research directions, it sets the stage for advancements in CBIR that will enhance retrieval accuracy, usability, and effectiveness in various application domains. © 2023, CC BY.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Seeing the Unseen: Learning Basis Confounder Representations for Robust Traffic Prediction 25

Seeing the Unseen: Learning Basis Confounder Representations...

引用

Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and data mining V.1

作者： Jiahao Ji Wentao Zhang Jingyuan Wang Chao Huang School of Computer Science and Engineering Beihang University Beijing China SCSE Beihang University Beijing China & MIIT Key Laboratory of Data Intelligence and Management SEM Beihang University Beijing China Department of Computer Science Musketeers Foundation Institute of Data Science University of Hong Kong Hong Kong SAR China

ISBN: (纸本)9798400712456

Traffic prediction is essential for intelligent transportation systems and urban computing. It aims to establish a relationship between historical traffic data X and future traffic states Y by employing various statistical or deep learning methods. However, the relations of X → Y are often influenced by external confounders that simultaneously affect both X and Y, such as weather, accidents, and holidays. Existing deep-learning traffic prediction models adopt the classic front-door and back-door adjustments to address the confounder issue. However, these methods have limitations in addressing continuous or undefined confounders, as they depend on predefined discrete values that are often impractical in complex, real-world scenarios. To overcome this challenge, we propose the Spatial-Temporal sElf-superVised confoundEr learning (STEVE) model. This model introduces a basis vector approach, creating a base confounder bank to represent any confounder as a linear combination of a group of basis vectors. It also incorporates self-supervised auxiliary tasks to enhance the expressive power of the base confounder bank. Afterward, a confounder-irrelevant relation decoupling module is adopted to separate the confounder effects from direct X → Y relations. Extensive experiments across four large-scale datasets validate our model's superior performance in handling spatial and temporal distribution shifts and underscore its adaptability to unseen confounders. Our model implementation is available at https://***/bigscity/STEVE_CODE.

关键词： continuous and undefined confounder

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Parallel Harris Hawks Optimization Based WSN Node Localization Algorithm

Journal of Network Intelligence

引用

Journal of Network Intelligence 2021年第4期6卷 688-703页

作者： He, Xiankang Chu, Shu-Chuan Liu, Shi-Jian Pan, Jeng-Shyang Yan, Lijun College of Computer Science and Mathematics Fujian University of Technology Fuzhou350118 China Shenzhen Institute of Information Technology Shenzhen518000 China College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China College of Science and Engineering Flinders University Adelaide Australia Fujian Provincial Key Laboratory of Big Data Mining and Applications Fujian University of Technology Fuzhou350118 China

The localization of wireless sensor network (WSN) is an increasingly promi-nent problem. The goal of this problem is to use the anchor nodes in WSN to esti-mate the geographical location of the unknown nodes. This paper proposes a novel al-gorithm, named dynamic parallel Harris Hawks optimization (DPHHO). It contains the dynamic control strategy of the escaping energy and the parallel communication mecha-nism. DPHHO algorithm significantly improves the global search ability of the original Harris Hawks optimization (HHO) algorithm. Also, a novel localization algorithm based on hop distance correction and DPHHO is proposed. The proposed DPHHO algorithm was tested on the 23 classical test functions and the DPHHO-DV-Hop algorithm is applied to the localization of WSN. The experimental results show that, compared with HHO and other optimization algorithms, the proposed DPHHO algorithm is more effective and efficient. Compared with DV-Hop and other localization algorithms, the proposed DPHHO-DV-Hop is an effective algorithm for the localization of WSN. © 2021, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Sensor nodes

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：