检索结果-内蒙古大学图书馆

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Wang, Quan Zhang, Licheng Guo, Zikang Mao, Zhendong MOE Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications Beijing China University of Science and Technology of China Hefei China

ISBN: (纸本)9782493814104

The effective detection of AI-generated text is a vital principle to ensure responsible use of large language models (LLMs). Previous studies mainly focused on discovering and utilizing internal evidences contained in the text itself to perform the detection, while ignoring external evidences implicated in an established knowledge graph (KG) which may also be key discriminative factors between AI-generated and human-written text. To address this deficiency, we propose IDEATE, a novel hierarchical graph network that utilizes both internal and external factual structures to detect AI-generated text. IDEATE consists of a mention-level subgraph at the bottom to describe internal factual structures of mentioned entities reflected in the input text, and an entity-level subgraph at the top to describe external factual structures of mentioned entities reflected in an external KG. Hierarchical graph convolution is then applied successively on the two subgraphs, through which the two types of factual structures will be embedded into the output and used for the final detection. Extensive experiments on four benchmarking datasets show that IDEATE consistently outperforms current state-of-the-art methods in detecting text generated by various LLMs, ranging from GPT-2 to the more powerful ChatGPT, verifying the necessity and superiority of introducing external evidences for AI-generated text detection. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Joint cache placement and beamforming design for scalable videos in C-RAN 15

Joint cache placement and beamforming design for scalable vi...

引用

15th International Symposium on Wireless Communication Systems, ISWCS 2018

作者： Ma, Mingfeng Lv, Tiejun Zhang, Xuewei Key Laboratory of Trustworthy Distributed Computing and Service Ministry of Education School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China

ISBN: (纸本)9781538650059

This paper investigates the joint optimization of layered caching and beamforming design for scalable videos delivered by Dynamic Adaptive Streaming over HTTP in the multicell cloud radio access network. To transmit differential scalable videos, each video is divided into a base layer and a set of enhancement layers based on scalable video coding. Taking the layered property of scalable videos into account, we propose a layered caching scheme, i.e., multiple layers of different scalable videos are cached in the baseband unit pool and the cache-enabled cluster respectively, which alleviates the burden of backhaul links effectively. Since the quality of videos received by each user depends on the available bandwidth, the appropriate scalable videos are provided for users based on the transmission rates. To provide superior video quality for users, we aim to maximize the sum rate for users accessing the remote radio heads and regard the backhaul consumption as a quality of service constraint explicitly. Then, the weighted sum rate maximization problem is formulated, which can be converted into a quadratically constrained quadratic programming problem through the l0-norm approximation technique and the weighted minimum mean square error approach. Finally, simulation results demonstrate the superiority of the proposed scheme. © 2018 IEEE.

关键词： Scalable video coding

来源：评论

学校读者我要写书评

暂无评论

VMScan: an out-of-VM malware scanner

引用

The Journal of China Universities of Posts and Telecommunications 2020年第4期27卷 59-68页

作者： Lin Jie Liu Chuanyi Fang Binxing School of Computer Science and Technology Harbin Institute of TechnologyShenzhenShenzhen 518055China Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and TelecommunicationsBeijing 100876China Department of Information and Electronic Engineering Chinese Academy of EngineeringBeijing 100088China

The harm caused by malware in cloud computing environment is more and more serious. Traditional anti-virus software is in danger of being attacked when it is deployed in virtual machine on a large scale, and it tends not to be accepted by tenants in terms of performance. In this paper, a method of scanning malicious programs outside the virtual machine is proposed, and the prototype is implemented. This method transforms the memory of the virtual machine to the host machine so that the latter can access it. The user space and kernel space of virtual machine memory are analyzed via semantics, and suspicious processes are scanned by signature database. Experimental results show that malicious programs can be effectively scanned outside the virtual machine, and the performance impact on the virtual machine is low, meeting the needs of tenants.

关键词： security virtualization cloud malware virus detection signature scanning

来源：评论

学校读者我要写书评

暂无评论

LRChain:data protection and sharing method of learning archives based on consortium blockchain

引用

The Journal of China Universities of Posts and Telecommunications 2024年第4期31卷 28-42页

作者： Lan Lina Gao Yuhan Shi Ruisheng Wu Fenfen School of Humanities Beijing University of Posts and TelecommunicationsBeijing 100876China School of Cyberspace Security Beijing University of Posts and TelecommunicationsBeijing 100876China Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and TelecommunicationsMinistry of EducationBeijing 100876China

Learning archives management in traditional systems faces challenges such as inadequate security,weak tamper resistance,and limited sharing *** address these issues,this paper proposes LRChain,a method based on consortium blockchain,for lifelong learning archives data protection and *** employs a combination of on-chain and off-chain cooperative storage using a consortium chain and inter planetary file system(IPFS)to enhance data security and *** also enables fine-grained verification of learning archives through selective disclosure principles,ensuring privacy protection of sensitive ***,an attributebased encryption(ABE)algorithm is utilized to establish authorized access control for learning archives,facilitating safe and trusted *** evaluations and security analyses demonstrate that this method exhibits decentralization,strong security,tamper resistance,and performs well,effectively meeting the requirements for secure sharing of learning archive data.

关键词： learning archives consortium blockchain collaborative storage secure sharing

来源：评论

学校读者我要写书评

暂无评论

Incorporate Personas and Emotions into Dialog Generation 5

Incorporate Personas and Emotions into Dialog Generation

引用

5th International Academic Exchange Conference on Science and Technology Innovation, IAECST 2023

作者： Wu, Yufeng Liu, Chengyu Guo, Wenjian Beijing University of Posts and Telecommunications Key Laboratory of Trustworthy Distributed Computing and Service Beijing China Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9798350357738

Since the Turing test, creating believable open-domain dialogues has been a basic problem in the study of artificial intelligence. Achieving success with dialogue models can greatly improve natural human-computer interactions and open up new possibilities for applications such as chatbots for therapy. Incorporating emotional information into dialogue systems and giving them personas can enhance user satisfaction by fostering more natural communication and human-like conversations. Both of these study directions are comparatively well-established, with important accomplishments in them. On the other hand, various roles in real-life discussion scenarios portray emotions in different ways. In order to create more fulfilling and human-like discussions, we investigate in this study the integration of personas and emotions in dialogue creation systems. Through the combination of explicit and implicit persons, we tackle the problem of individualized and emotional conversation. While implicit personas are gleaned from linguistic style, colloquial idioms, and personality attributes, explicit personas are derived from user profiles. Explicit personas and emotional labels are incorporated into the dialogue agent, and implicit personas are captured through the use of latent variables in a pre-trained Transformer model. Our suggested method enhances the authenticity of discussions and user experience by producing responses that are consistent with personas and emotions, hence improving the quality of communication. © 2023 IEEE.

关键词： Human computer interaction

来源：评论

学校读者我要写书评

暂无评论

TKCA:a timely keystroke-based continuous user authentication with short keystroke sequence in uncontrolled settings

引用

Cybersecurity 2021年第1期4卷 177-192页

作者： Lulu Yang Chen Li Ruibang You Bibo Tu Linghui Li Institute of Information Engineering Chinese Academy of Sciences100093 BeijingChina School of Cyber Security University of Chinese Academy of Sciences100049 BeijingChina Key Laboratory of Trustworthy Distributed Computing and Service Ministry of EducationBeijing University of Posts and Telecommunications100876 BeijingChina

keystroke-based behavioral biometrics have been proven effective for continuous user *** state-of-the-art algorithms have achieved outstanding results in long text or short text collected by doing some *** remains a considerable challenge to authenticate users continuously and accurately with short keystroke inputs collected in uncontrolled *** this work,we propose a Timely keystroke-based method for Continuous user Authentication,named *** integrates the key name and two kinds of timing features through an embedding *** it captures the relationship between context keystrokes by the Bidirectional Long Short-Term Memory(Bi-LSTM)*** conduct a series of experiments to validate it on a public dataset-the Clarkson II dataset collected in a completely uncontrolled and natural *** results show that the proposed TKCA achieves state-of-the-art performance with 8.28%of EER when using only 30 keystrokes and 2.78%of EER when using 190 keystrokes.

关键词： keystroke dynamics Continuous user authentication Embedding LSTM Bi-LSTM

来源：评论

学校读者我要写书评

暂无评论

Enhanced mutual authentication model of IoT

引用

Journal of China Universities of Posts and Telecommunications 2013年第SUPPL -2期20卷 69-74页

作者： Yang, Jin-Cui Pang, Hao Zhang, Xin School of Software Engineering Beijing University of Posts and Telecommunications Beijing 100876 China Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications Beijing 100876 China Bank of Communications Beijing Branch Beijing 100032 China

The authentication mechanism of the Internet of things (IoT) control system is an important research field. This paper aims to give an appropriate authentication mechanism of IoT control system. It analyzes the pros and cons of some existing authentication mechanisms, proposes an enhanced bi-direction authentication mechanism for IoT control system, discusess the proposed mechanism in detail, including the improvements measures, the authentication process and the authentication model. Finally, this paper presents the security analysis of the enhanced authentication model. Analysis results indicate that the model is a relatively feasible security authentication model for the IoT control system. © 2013 The Journal of China Universities of Posts and Telecommunications.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Web Attack Detection through Network-Traffic-Based Feature Engineering and Machine Learning

Web Attack Detection through Network-Traffic-Based Feature E...

引用

2020 International Conference on Cyberspace Innovation of Advanced Technologies, CIAT 2020

作者： Yang, Jian Wang, Hao Lu, Yueming School of Cyberspace Security Key Lab of Trustworthy Distributed Computing and Service Ministry of Education Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9781450387828

The design of feature set is a basic research direction of anomaly detection technology. The characterization ability of feature set directly affects the accuracy and generalization ability of intrusion detection algorithm. KDD CUP 99 is the most widely used data set in anomaly detection related research papers, but the feature set needs to analyze the payload content characteristics of packets, so the feature set cannot be applied to the encrypted traffic. CICIDS2017 is also a widely used data set. However, because data in PCAP format can only be used as input, the generalization ability of the model trained by this data set is poor when it is used for real-time attack detection. Based on the above problems, a feature set based on network flow characteristics is designed and implemented in this paper, which includes the time-window-based statistical features related to network traffic and host-related statistical features. Experiments show that the anomaly detection model trained by the feature set can meet the requirements of both real-time and classification accuracy. © 2020 ACM.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

A recommend system for modelling large-scale advertising

A recommend system for modelling large-scale advertising

引用

International Conference on Cyberspace Technology, CCT 2014

作者： Ma, Jing Lu, Yueming School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China Key Laboratory of Trustworthy Distributed Computing and Service BUPT Ministry of Education Beijing China

ISBN: (纸本)9781849199285

In a large-scale network environment, the Internet has produced an amount of digital information such as social media, e-commercial comments, online shopping records and advertising click logs. So, recommend systems not only furnish users with satisfied requirements but also bring about healthy profit to companies. To cope with massive data, we proposed an effective recommend system (ERS) for modelling abundant history data. Firstly, we advise a new data split strategy to make the recommend system fit for large scale advertising. Then, a distributed-recommendation algorithm is designed. Finally, we apply ERS and distributed computing and storage tools (such as Spark, Hadoop, etc.) to model history data aiming at improving the system efficiency. In this paper, we also compare and analyse diverse performances of ERS including its efficiency and its accuracy.

关键词： Metadata

来源：评论

学校读者我要写书评

暂无评论

A method based on MD5 and time for preventing deception in electronic commerce

A method based on MD5 and time for preventing deception in e...

引用

International Conference on Cyberspace Technology, CCT 2014

作者： Hu, Zhenggang Lu, Yueming School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China Key Laboratory of Trustworthy Distributed Computing and Service BUPT Ministry of Education Beijing China

ISBN: (纸本)9781849199285

Aiming at reducing the risk of data being intercepted and preventing deceptions in the electronic commerce, this paper proposes a method based on MD5 and time which can effectively prevent the deception in data transmission. The method uses the traditional MD5 algorithm to encrypt and validate important data, then verifying the validity and legitimacy of the data by using time.

关键词： Electronic commerce MD5 algorithm Preventing deception Time

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：