检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Wang, Yaxian Ding, Henghui He, Shuting Jiang, Xudong Wei, Bifan Liu, Jun School of Computer Science and Technology Xi'an Jiaotong University China Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi'an Jiaotong University China Institute of Big Data Fudan University China Shanghai University of Finance and Economics China Nanyang Technological University Singapore School of Continuing Education Xi'an Jiaotong University China Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi'an Jiaotong University China

In this work, we address the challenging task of Generalized Referring Expression Comprehension (GREC). Compared to the classic Referring Expression Comprehension (REC) that focuses on single-target expressions, GREC extends the scope to a more practical setting by further encompassing notarget and multi-target expressions. Existing REC methods face challenges in handling the complex cases encountered in GREC, primarily due to their fixed output and limitations in multi-modal representations. To address these issues, we propose a Hierarchical Alignment-enhanced Adaptive Grounding network (HieA2G) for GREC, which can flexibly deal with various types of referring expressions. First, a Hierarchical Multi-modal Semantic Alignment (HMSA) module is proposed to incorporate three levels of alignments, including word-object, phrase-object, and text-image alignment. It enables hierarchical cross-modal interactions across multiple levels to achieve comprehensive and robust multi-modal understanding, greatly enhancing grounding ability for complex cases. Then, to address the varying number of target objects in GREC, we introduce an Adaptive Grounding Counter (AGC) to dynamically determine the number of output targets. Additionally, an auxiliary contrastive loss is employed in AGC to enhance object-counting ability by pulling in multi-modal features with the same counting and pushing away those with different counting. Extensive experimental results show that HieA2G achieves new state-of-the-art performance on the challenging GREC task and also the other 4 tasks, including REC, Phrase Grounding, Referring Expression Segmentation (RES), and Generalized Referring Expression Segmentation (GRES), demonstrating the remarkable superiority and generalizability of the proposed HieA2G. Copyright © 2025, The Authors. All rights reserved.

关键词： Electric grounding

来源：评论

学校读者我要写书评

暂无评论

HyperDiff: Masked Diffusion Model with High-efficient Transformer for Hyperspectral Image Cross-Scene Classification

HyperDiff: Masked Diffusion Model with High-efficient Transf...

引用

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

作者： Pei Zhang Dong Wang Chanyue Wu Jing Yang Lei Kang Zongwen Bai Ying Li Qiang Shen School of Computer Science Northwestern Polytechnical University Xi’an China Shaanxi Key Laboratory of Intelligent Processing for Big Energy Data Yan’an University Yan’an China School of Automation and Software Engineering Shanxi University Taiyuan China Qingdao Topscomm Communication Co. Ltd Qingdao China Department of Computer Science Aberystwyth University UK

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Hyperspectral Image (HSI) cross-scene classification is a challenging task in remote sensing, particularly when real-time processing of Target Domain (TD) HSI is required, and data cannot be reused for training. While deep learning methods have shown promising results, the generalization ability of HSI representations remains limited, mainly due to class label imbalance. This paper introduces a dual-stage learning framework based on transfer learning to enhance classification accuracy in the TD. The framework includes a self-supervised learning stage and a supervised fine-tuning stage. The self-supervised stage focuses on learning robust representations by leveraging inherent structures within HSI data, while the fine-tuning stage uses training labels to extract semantic information. A masked diffusion model predicts masked tokens from unmasked ones, capturing both high-level structures and fine details in HSI data. An efficient spatiospectral Transformer, which removes self-attention from the decoder, is proposed to enhance the self-supervised process. This design allows mask tokens to obtain information from visible tokens without interacting with each other, reducing sequence length and computational costs. By decoding each mask token conditionally independently, only a subset of masked tokens is processed. Extensive experiments on two public HSI datasets demonstrate that the proposed method outperforms state-of-the-art techniques.

关键词： Training Transfer learning Self-supervised learning Transformers Diffusion models Real-time systems Decoding Computational efficiency Speech processing Hyperspectral imaging

来源：评论

学校读者我要写书评

暂无评论

Density PSO-based software module clustering algorithm

引用

The Journal of China Universities of Posts and Telecommunications 2018年第4期25卷 38-47页

作者： Sun Jiaze Ling Beilei School of Computer Science and Technology Xi'an University of Posts and Telecommunications Shaanxi Key Laboratory of Network Data Intelligent Processing Xi'an University of Posts and Telecommunications

Software module clustering is to divide the complex software system into many subsystems to enhance the intelligibility and maintainability of software systems. To increase convergence speed and optimize clustering solution,density PSO-based( DPSO) software module clustering algorithm is proposed. Firstly,the software system is converted into complex network diagram,and then the particle swarm optimization( PSO) algorithm is *** shortest path method is used to initialize the swarm,and the probability selection approach is used to update the particle positions. Furthermore,density-based modularization quality( DMQ) function is designed to evaluate the clustering quality. Five typical open source projects are selected as benchmark programs to verify the efficiency of the DPSO algorithm. Hill climbing( HC) algorithm,genetic algorithm( GA),PSO and DPSO algorithm are compared in the modularization quality( MQ) and DMQ. The experimental results show that the DPSO is more stable and more convergent than the other three traditional algorithms. The DMQ standard is more reasonable than MQ standard in guiding software module clustering.

关键词： software module clustering complex network PSO MQ modularity density

来源：评论

学校读者我要写书评

暂无评论

An attention-based cascade R-CNN model for sternum fracture detection in X-ray images

引用

CAAI Transactions on Intelligence Technology 2022年第4期7卷 658-670页

作者： Yang Jia Haijuan Wang Weiguang Chen Yagang Wang Bin Yang School of Computer Xi'an University of Posts and TelecommunicationsXi'anShaanxiChina Shaanxi Key Laboratory of Network Data Intelligent Processing Xi'an University of Posts and TelecommunicationsXi'anShaanxiChina Xi'an Key Laboratory of Big Data and Intelligent Computing Xi'anShaanxiChina Department of Radiology Xi'an Honghui HospitalXi'anChina

Fracture is one of the most common and unexpected *** not treated in time,it may cause serious consequences such as joint stiffness,traumatic arthritis,and nerve *** computer vision technology to detect fractures can reduce the workload and misdiagnosis of fractures and also improve the fracture detection ***,there are still some problems in sternum fracture detection,such as the low detection rate of small and occult *** this work,the authors have constructed a dataset with 1227 labelled X-ray images for sternum fracture *** authors designed a fully automatic fracture detection model based on a deep convolution neural network(CNN).The authors used cascade R-CNN,attention mechanism,and atrous convolution to optimise the detection of small fractures in a large X-ray image with big local *** authors compared the detection results of YOLOv5 model,cascade R-CNN and other state-of-the-art *** authors found that the convolution neural network based on cascade and attention mechanism models has a better detection effect and arrives at an mAP of 0.71,which is much better than using the YOLOv5 model(mAP=0.44)and cascade R-CNN(mAP=0.55).

关键词： attention mechanism cascade R-CNN fracture detection X-ray image

来源：评论

学校读者我要写书评

暂无评论

Edge Computing Task Scheduling with Joint Blockchain and Task Caching in Industrial Internet

引用

Computers, Materials & Continua 2023年第4期75卷 2101-2117页

作者： Yanping Chen Xuyang Bai Xiaomin Jin Zhongmin Wang Fengwei Wang Li Ling School of Computer Science and Technology Xi’an University of Posts and TelecommunicationsXi’an710121China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi’an710121China Xi’an Key Laboratory of Big Data and Intelligent Computing Xi’anShaanxi710121China ZTE Corporation Shenzhen51805China

Deploying task caching at edge servers has become an effectiveway to handle compute-intensive and latency-sensitive tasks on the industrialinternet. However, how to select the task scheduling location to reduce taskdelay and cost while ensuring the data security and reliable communicationof edge computing remains a challenge. To solve this problem, this paperestablishes a task scheduling model with joint blockchain and task cachingin the industrial internet and designs a novel blockchain-assisted cachingmechanism to enhance system security. In this paper, the task schedulingproblem, which couples the task scheduling decision, task caching decision,and blockchain reward, is formulated as the minimum weighted cost problemunder delay constraints. This is a mixed integer nonlinear problem, which isproved to be nonconvex and NP-hard. To solve the optimal solution, thispaper proposes a task scheduling strategy algorithm based on an improvedgenetic algorithm (IGA-TSPA) by improving the genetic algorithm initializationand mutation operations to reduce the size of the initial solutionspace and enhance the optimal solution convergence speed. In addition,an Improved Least Frequently Used algorithm is proposed to improve thecontent hit rate. Simulation results show that IGA-TSPA has a faster optimalsolution-solving ability and shorter running time compared with the existingedge computing scheduling algorithms. The established task scheduling modelnot only saves 62.19% of system overhead consumption in comparison withlocal computing but also has great significance in protecting data security,reducing task processing delay, and reducing system cost.

关键词： Edge computing task scheduling blockchain task caching industrial security

来源：评论

学校读者我要写书评

暂无评论

SolBERT: Advancing solidity smart contract similarity analysis via self-supervised pre-training and contrastive fine-tuning

引用

Information and Software Technology 2025年 184卷

作者： Zhenzhou Tian Yudong Teng Xianqun Ke Yanping Chen Lingwei Chen School of Computer Science and Technology Xi’an University of Posts and Telecommunications Xi’an 710121 Shaanxi China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi’an University of Posts and Telecommunications Xi’an 710121 Shaanxi China Xi’an Key Laboratory of Big Data and Intelligent Computing Xi’an University of Posts and Telecommunications Xi’an 710121 Shaanxi China Department of Computer Science and Engineering Wright State University Dayton 45435 OH USA

Context: Reliable and effective similarity analysis for the smart contracts facilitates the maintenance and quality assurance of the smart contract ecosystem. However, existing signature-based methods and code representation learning-based methods suffer from limitations such as heavy-weight program analysis payloads or suboptimal contract encodings. Objective: This paper aims to design a fully unsupervised language model for better capturing the syntactic and semantic richness of Solidity code, and utilizes it for advancing the effectiveness of smart contract similarity analysis. Methods: Inspired by the impressive semantic learning capability of pre-trained language models (PLMs), we propose SolBERT, a PLM specifically tailored for enhancing Solidity smart contracts similarity detection. To ensure it produces high-quality encodings, SolBERT leverages BERT-style pre-training with the masked language modeling (MLM) and token type prediction (TTP) tasks applied on code-structure-aware token sequences derived from the contracts’ abstract syntax trees (ASTs) through structure-retaining tree linearization and light-weight normalization to learn a base model. On this basis, self-supervised contrastive fine-tuning and unsupervised whitening operations are further performed to optimize contract encoding generation. Results: Experiments are conducted on three contract similarity-related tasks, including contract clone detection, bug detection, and code clustering. The results indicate that SolBERT significantly outperforms state-of-the-art approaches with average absolute gains of 21.33% and 21.50% in terms of F1, and 17.78% and 26.60% in terms of accuracy for the clone detection and bug detection tasks, respectively; and an average absolute gain of 17.97% for code clustering task. When applying both contrastive fine-tuning and whitening optimizations, SolBERT also shows superior performance than the case of lacking any of them. Conclusion: The proposed approach, SolBERT, c

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multimodal Sentiment Analysis of Graphic Texts Based on Multicategorical Relative Fusion 6

Multimodal Sentiment Analysis of Graphic Texts Based on Mult...

引用

6th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2023

作者： Zhai, Sheping Gao, Mingyang Yang, Rui Kang, Xinnian Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an University of Posts and Telecommunications China

ISBN: (纸本)9798400707674

Unlike the traditional unimodal sentiment analysis, multimodal sentiment analysis can jointly process features between different modalities. Existing multimodal sentiment analysis methods simply extract features from text and images ignoring the influence of emoticons that may be contained in text and scene information in images on sentiment polarity judgment. At the same time, the unimodal feature fusion stage does not fully fuse the features of each modality resulting in low accuracy of sentiment analysis. Therefore, this paper proposes a multimodal sentiment analysis model based on multicategorical relative fusion of images and texts. which separates text content and emoji in the word embedding stage in the feature extraction stage, the traditional emoji sentiment dictionary is queried to obtain their sentiment scores and attention mechanism is introduced to extract scene features and object features of images respectively;in the feature fusion stage, the emoji sentiment scores. mixed image. The final multimodal features are obtained by fusion the emoji sentiment scores. mixed image features and text features with layer normalization. The experiments on the dataset Yelp show that the improvement of the model in the feature extraction and feature fusion stages effectively improves the accuracy and F1 value of multimodal sentiment analysis in which the accuracy increases by 4.7 percentage points on the New York dataset and the F1 value reaches 85.7% on the San Francisco dataset. © 2023 ACM.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Moving Forward Together: A Multi-User Pattern Extraction Neural Model for Sequential Recommendations 6

Moving Forward Together: A Multi-User Pattern Extraction Neu...

引用

6th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2023

作者： Du, Xiangpeng Li, Xiaoge Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xian University of Posts and Telecommunications Shaanxi Xi'an710121 China

ISBN: (纸本)9798400707674

The sequential recommendation is a very important task in recommendation systems. The aim of it is to dynamic predict user's interests based on their historical behaviors. Despite recent progress, most of deep learning approaches focus on extracting the behavioral patterns of a single user. Pulling multiple users can increase the diversity of historical information. In this paper, we propose multi-user behavioral pattern extraction model (TMGN) via a combination of traditional behavioral pattern extraction and graph convolutional aggregation. Concretely, TMGN first learns the graph structure of users and items. Then it performs graph convolution aggregation to derive a representation of users to extract user interest features. We recall the users with the highest similarity by calculating the cosine similarity of each candidate user. Finally, we inject multiple users' sequence patterns and time information into the muti-head attention mechanism. Experiments on actual data (Movielens1m, Movielen20m and Amazon beauty) show that the proposed method outperforms current state-of-the-art sequential recommendation methods. TMGN achieves an average improvement of 5.67%, 5.75%, 7.32%, and 7.19% over the most robust baseline in Hit@5, Hit@10, NDCG@5, and NDCG@10, respectively. We provide extensive ablation studies to analyze the impact of each model component on TMGN. © 2023 ACM.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Application study of oil pipe entity recognition combined with Deep Reinforcement Learning: Named Entity Recognition combined with Deep Reinforcement Learning 6

Application study of oil pipe entity recognition combined wi...

引用

6th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2023

作者： Guan, Wei Liu, Shijie Ma, Li Xi'an University of Posts and Telecommunications Shaanxi Provincial Key Laboratory of Network Data Analysis and Intelligent Processing China

ISBN: (纸本)9798400707674

Aiming at the problems of inaccurate extraction of entity feature information and mislabeling of recognition results in the construction of knowledge graphs in the field of oil pipe failure using traditional methods, this paper proposes a research method of BERT-BiLSTM-CRF named entity recognition combined with deep reinforcement learning. The method firstly uses the BERT (Bidirectional encoder representations from transformers) pre-training model to obtain the word vectors of the input sequence semantics;secondly, the trained word vectors are fed into the bi-directional long short- term memory network (BiLSTM) to further obtain contextual features and output the maximum probability sequence annotation results according to the labeling rules of conditional random fields (CRF) and sequence decoding ability;finally, the BERT-BiLSTM-CRF model is supplemented with the deep reinforcement learning-based labeling Finally, the BERT-BiLSTM-CRF-DRL model is added to the BERT-BiLSTM-CRF model to form the framework of named entity recognition model in oil pipe failure domain. The model is compared with other 2 named entity recognition models on a self-built oil pipe data-set, and the experimental results show that the accuracy, recall and F1 values of the model reach 90.5%, 92.7% and 91.9%, respectively, and the entity recognition is better than the other 2 models. © 2023 ACM.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Visual Question Answering Model Based on Relationship graphs and Co-Attention 6

Visual Question Answering Model Based on Relationship graphs...

引用

6th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2023

作者： Guan, Wei Yang, Xueli Ma, Li Xi'an University of Posts and Telecommunications Shaanxi Provincial Key Laboratory of Network Data Analysis and Intelligent Processing China

ISBN: (纸本)9798400707674

Visual Question Answering is a very challenging tasks in the fiield of AI. The existing model is not accurate enough to judge the relationship between problem words, image objects and the association between problem and image objects, meanwhile, the interpretability of the model needs to be improved. In response, this paper proposes a visual question answering model based on relationship graph and co-attention. Relational features can be obtained by constructing a relational graph using the relationships between the problem and the image, and between the objects of the image, and by dynamically updating the relational graph using a graph attention network guided by the problem. Then, the obtained features are fused using co- attention to obtain richer ***, these features are used to make answer prediction. The model is experimented on the VQA2.0 dataset, and the experimental results show that the model improved the accuracy of the visual question answering model to some extent compared with the existing models. © 2023 ACM.

关键词： Image enhancement

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：