检索结果-内蒙古大学图书馆

4th IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2024

作者： Cao, Jiaxi Zhang, Chunjie Zheng, Xiaolong Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology School of Computer Science and Technology Beijing China Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Beijing China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing China

ISBN: (纸本)9798350349252

Dense video captioning, a task that has recently garnered attention in the field of computer vision, focuses on generating detailed captions for multiple events along with their temporal locations within untrimmed videos. Existing transformer-based methods have achieved notable success. However, some methods utilizing deformable transformer face challenges in processing long videos, as the encoder may struggle to capture the long-term dependencies of events in the video. This issue is partly due to not integrating Convolutional Neural networks (CNN) with the deformable transformer architecture. In this work, we introduce Dense Video Captioning through Convolutional-Transformer Integration (CTDVC), which links Layer Normalization, the Scalable-Granularity Perception (SGP), and Group Normalization into a network, replacing the encoder part of the deformable transformer. This model better accounts for the sequential relationship between events, especially those events in the video that span over a longer duration, allowing it to generate video captions that are not only more coherent but also more precise in their narrative. We evaluated our method on ActivityNet Captions and YouCook2 datasets. Experiments demonstrate that our proposed method CTDVC achieves an improvement in performance. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

HIFI-Stego: A High-Fidelity Embedding Audio Steganography Based on Audio Features Decoupling

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 2032-2044页

作者： Sanfeng Zhang Baiyu Tian Yang Gao Xingyu Liu Wang Yang School of Cyber Science and Engineering Southeast University Nanjing China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing China

The higher audio quality of steganography is directly correlated with the increased resistance to steganalysis tools. The advancement of generative AI technologies, particularly those that decouple style features from content, has shown promising developments by facilitating the creation of superior media content. This paper introduces the concept of audio decoupling and presents HIFI-Stego, an embedding audio steganography technique that aims to improve security while maintaining elevated stego audio quality. HIFI-Stego comprises a generator based on the encoder-decoder architecture and a secret message extractor. The encoder of the generator decouples the original audio, yielding the content vector, while the vocoder WORLD is employed to extract the style vector to preserve high-quality audio related features such as timbre and tone. Subsequently, the decoder embeds the secret vector into the decoupled content vector and then couples it with the style vector to generate high-fidelity stego audio. As embedding is not done in traditional time domain or frequency domain, existing analysis tools targeted at traditional steganographies fail to effectively detect the presence of the hidden message. The secret message extractor reuses the generator encoder and augments it with a single-layer convolutional neural network, resulting in a simplified structure suitable for lightweight deployment. Experimental results demonstrate that HIFI-Stego outperforms traditional generative and embedding steganographies in terms of audio quality, steganographic capacity, anti-analysis ability, and concealment.

关键词： Steganography Vectors Time-domain analysis Generators Feature extraction Training Resistance Generative adversarial networks Convolutional neural networks Security

来源：评论

学校读者我要写书评

暂无评论

Adversarial attack method with adaptive step size for ID card text recognition 4

Adversarial attack method with adaptive step size for ID car...

引用

4th IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2024

作者： Feng, Weidong Zhang, Chunjie Zheng, Xiaolong Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology School of Computer Science and Technology Beijing China Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Beijing China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing China

ISBN: (纸本)9798350349252

In the field of text recognition models, existing adversarial attack algorithms mainly target white-box scenarios, which are usually limited to single-color backgrounds and short text length. However, with the complex color textures of ID card images and long text length, these methods are not satisfactory when performing adversarial attacks on ID card text data. To overcome these challenges, this paper introduces a novel black-box attack method tailored for ID card text recognition models. The proposed approach accounts for the complexities of intricate image backgrounds, extended text data, and the limitations inherent in black-box conditions. The proposed algorithm introduces an adaptive step-size search method based on line search, aiming to overcome the limitations of traditional methods that rely on fixed step sizes and projection techniques for perturbation generation. This refinement aims to enhance the success rate of black-box attacks while maintaining efficacy in white-box scenarios. Spatial transformations and random noise are incorporated to fortify the robustness of adversarial samples. Experimental results, conducted under various norms, demonstrate that the fusion of the traditional method with the adaptive step-size search method yields superior attack performance. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

STA+: Spatiotemporal Adaptation with Adaptive Model Selection for Video Action Recognition 4

STA+: Spatiotemporal Adaptation with Adaptive Model Selectio...

引用

4th IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2024

作者： Li, Minghui Zhang, Chunjie Zheng, Xiaolong Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology School of Computer Science and Technology Beijing China Chinese Academy of Sciences State Key Laboratory of Multimodal Artifical Intelligence Systems Institute of Automation Beijing China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing China

ISBN: (纸本)9798350349252

Recent breakthroughs in video models have achieved remarkable success by integrating vision transformers into the video domain through adaptation. However, prevalent approaches in existing literature often entail a single branch, incorporating additional spatiotemporal learning modules. This approach cannot fully satisfy the unique and targeted learning requirements of temporal and spatial representations. It lacks a well-balanced spatiotemporal understanding of videos, particularly in recognizing human actions, which requires a comprehensive understanding of space and time. In this study, we introduce a novel two-stream architecture named SpatioTemporal Adaptation with Adaptive Model Selection (STA+) to achieve synchronized spatiotemporal understanding of video with only RGB inputs. We propose an adaptive model selection method based on the intrinsic categorization judgment of the samples to enhance overall performance. We validate the effectiveness of our proposed method through rigorous experiments conducted on public benchmarks such as Diving-48 and Something-Something- V2. Our method consistently achieves competitive performance on these datasets. © 2024 IEEE.

关键词： adaptation spatiotemporal Video action recognition

来源：评论

学校读者我要写书评

暂无评论

The drivers and spatial distribution of economic efficiency in China's cities

引用

Journal of Geographical Sciences 2022年第8期32卷 1427-1450页

作者： CAO Yanni WU Tong KONG Lingqiao WANG Xuezhi ZHANG Lufeng OUYANG Zhiyun State Key Laboratory of Urban and Regional Ecology Research Center for Eco-Environmental SciencesCASBeijing 100085China University of Chinese Academy of Sciences Beijing 100049China Computer Network Information Center CASBeijing 100190China

This study analyzes cities in China at the prefecture level and above to calculate indices for“urban economic efficiency”(the relationship between input factors and output)and“urbanization economic efficiency”(the relationship between newly increased output and increased economic input),based on the Stochastic Frontier Analysis(SFA)*** compare and analyze the factors influencing change and their spatial *** results show that capital and labor rather than urban land could effectively improve urban and urbanization economic ***,although the proportion of wages to GDP has a significant negative impact on urban economic efficiency,for social equity and stability,the proportion should be increased;if appropriate,it would not significantly reduce urbanization economic ***,population density,population urbanization rate,and government fiscal expenditure significantly positively impact urban and urbanization economic ***,we also found that increases in the degree of industrial structure deviation and urban landscape fragmentation are harmful to urbanization economic *** terms of spatial distribution,the urbanization economic efficiency of most of China's northeastern and eastern coastal areas is significantly lower than that of other regions;at the same time,the urban economic efficiency of most of these cities has been decreasing,especially in the northeast,which warrants greater policy attention.

关键词： development planning economic development urban economics urban planning China

来源：评论

学校读者我要写书评

暂无评论

Research on long and short-term social recommendation based on convolutional and gated recurrent units 29

Research on long and short-term social recommendation based ...

引用

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Jia, Zihe Xue, Peng Dai, Zhiqiang Gao, Qian Zhang, Xiaomeng Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Jinan250014 China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Jinan250353 China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computer Networks Jinan250014 China

ISBN: (纸本)9798350330717

The development of the Internet has made people more closely related and has put forward higher requirements for recommendation models. Most recommendation models are studied only for the long-term interests of users. In this paper, the interaction time between the user and the item is introduced as auxiliary information in the model construction. Interaction time is used to determine users' long-term preferences and short-term preferences. In this paper, temporal features are extracted by building a convolutional gated recurrent unit with attention neural network (CNN-GRU-Attention). Firstly, for the problem of accurate feature extraction, CNN are constructed to extract higher-level and more abstract features of themselves and transform high-dimensional data into low-dimensional data;secondly, for the problem of social temporality, GRU are used to not only extract temporal information, but also effectively reduce gradient dispersion, making model convergence and training easier;finally, Graph Attention networks are used to aggregate the social relationship information of users and items respectively, which constitute the final feature representation of users and items respectively. In particular, a modified cosine similarity is used to reduce the error caused by data insensitivity when constructing the social information of the item. In this study, simulation experiments are conducted on two publicly available datasets (Epinions and Ciao), and the experimental results show that the proposed recommended model performs better than other social recommendation models, improving the evaluation metrics of MAE and RMSE by 1.06%-1.33% and 1.19%-1.37%, respectively. The effectiveness of the model innovation is proved. © 2023 IEEE.

关键词： convolutional gated recurrent unit graph attention networks neighborhood recommendation systems social networks

来源：评论

学校读者我要写书评

暂无评论

The Analysis of Schnorr Multi-Signatures and the Application to AI 1

引用

1st International Conference on Artificial Intelligence Security and Privacy, AIS and P 2023

作者： Wang, Wenchao Qin, Jing Liu, Jinlu Zhang, Xi Hou, Xinyi Wei, Zhongkai School of Mathematics Shandong University Jinan250100 China College of Computer and Cyber Security Hebei Normal University Shijiazhuang050024 China Hebei Provincial Key Laboratory of Network and Information Security Hebei Normal University Shijiazhuang050024 China

ISBN: (数字)9789819997855

ISBN: (纸本)9789819997848

Artificial intelligence (AI) is having a profound impact on our daily lives. We suggest using digital signatures to protect the user’s identity and achieve data accountability. To address high-risk applications, multi-signatures are expected to play an important role in AI. MuSig2 by Nick et al. is an efficient and secure Schnorr multi-signature scheme. MuSig2 implements signature aggregation and key aggregation, and MuSig2 is reduced to the One-More Discrete Logarithms (OMDL) problem in the random oracle model. This comes at the cost that the signer needs four nonces instead of one nonce for each signature. However, MuSig2 ignores the change of nonces in the forking lemma, which leads to the signer signature requiring too many nonces, and makes the proof of the scheme complicated. In this paper, we reduce the number of nonces from 4 to 2 and simplify the security proof of the MuSig2 scheme in the random oracle model. Then by reducing the security requirement slightly, we achieve the MuSig2 scheme’s security when the nonces are reused. Finally, we utilize the proof technology of MuSig2 to reduce the MSDL (Discrete-Logarithm based Multi-Signature) scheme by Boneh et al. to the OMDL assumption. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

GHTO: Improving Byzantine Robustness in Distributed Machine Learning Through Gradient Heterogeneity Twofold Optimization 6

GHTO: Improving Byzantine Robustness in Distributed Machine ...

引用

6th International Conference on Frontier Technologies of information and computer, ICFTIC 2024

作者： Li, Caiyun Yang, Ming Cao, Jian Wu, Xiaoming He, Yunpeng Wang, Xin Jinan250014 China Shandong Provincial Key Laboratory of Industrial Network and Information System Security Shandong Fundamental Research Center for Computer Science Jinan250014 China Shandong Evayinfo Technology Co. Ltd Jinan250101 China

ISBN: (纸本)9798331541750

Heterogeneous datasets induce local gradient variance and global gradient variance in distributed learning, collectively referred to as gradient heterogeneity. Gradient heterogeneity increases the system's vulnerability to Byzantine attacks, affecting performance. To address this issue, the paper proposes a Gradient Heterogeneity Twofold Optimization (GHTO) algorithm to enhance the Byzantine robustness of the system in non-independent and identically distributed environments. Specifically, a Local Gradient Correction (LGC) algorithm is proposed, which aims to reduce the local gradient variance. By further integrating LGC with momentum and bucketing techniques to achieve dual optimization of local and global gradient variance. Experimental results demonstrate the proposed LGC algorithm diminishes local gradient variance, and the GHTO method optimizes both local and global gradient variance, which further strengthens the robustness of the system against Byzantine attacks. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Visual malware classification and detection method based on convolutional neural network 9

Visual malware classification and detection method based on ...

引用

9th IEEE International Conference on Data Science in Cyberspace, DSC 2024

作者： Huang, Hua Sheng, Hua Han, Xingwang Zhang, Wenbin Zhao, Zilin State Grid Shandong Electric Power Company Information and Telecommunications Company Jinan China Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center National Supercomputer Center in Jinan Jinan China

ISBN: (纸本)9798350391367

With the rapid development of Internet technology, leading to the evolution of malware variants, and malicious threats are also increasing. Therefore, rapid and accurate classification of malware becomes crucial. In response to this challenge, this paper proposes a malware classification method based on deep learning and visualization techniques. The method first extracts sample features, then converts malware binary samples into image format, thus transforming abstract binary data into images with visual information. Subsequently, color mapping technique is applied to convert the processed images into color images. Furthermore, to address imbalanced datasets, this paper introduces data augmentation methods to ensure more balanced and comprehensive training of the model across different sample categories. Finally, an improved convolutional neural network architecture is employed for the classification detection of malware families, achieving accurate classification and identification of different malware families through learning and analysis of sample features. To evaluate the effectiveness of the proposed method, experiments are conducted on the Google Code Jam (GCJ) benchmark dataset. The experimental results demonstrate an accuracy rate of 99.29% on the GCJ dataset. Additionally, we also compared the performance of the improved method before and after, and found that the proposed method can provide more comprehensive feature representation and significantly improve classification accuracy and detection efficiency. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Self-Evaluation of Large Language Model based on Glass-box Features

Self-Evaluation of Large Language Model based on Glass-box F...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Huang, Hui Qu, Yingqi Liu, Jing Yang, Muyun Xu, Bing Zhao, Tiejun Lu, Wenpeng Faculty of Computing Harbin Institute of Technology Harbin China Baidu Inc. Beijing China Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology Jinan China

ISBN: (纸本)9798891761681

The proliferation of open-source Large Language Models (LLMs) underscores the pressing need for evaluation methods. Existing works primarily rely on external evaluators, focusing on training and prompting strategies. However, a crucial aspect - model-aware glass-box features - is overlooked. In this study, we explore the utility of glass-box features under the scenario of self-evaluation, namely applying an LLM to evaluate its own output. We investigate various glass-box feature groups and discovered that the softmax distribution serves as a reliable quality indicator for self-evaluation. Experimental results on public benchmarks validate the feasibility of self-evaluation of LLMs using glass-box features. © 2024 Association for Computational Linguistics.

关键词： Open systems

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：