检索结果-内蒙古大学图书馆

IEEE International Conference on Communications (ICC)

作者： Gaochang Xie Renchao Xie Xinyuan Zhang Jiangtian Nie Qinqin Tang Qian Chen Dusit Niyato State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China Purple Mountain Laboratories Nanjing China School of Computer Science and Engineering Nanyang Technological University Singapore Singapore State Key Laboratory of Integrated Service Networks Xidian University Xi'an China

ISBN: (数字)9781728190549

ISBN: (纸本)9781728190556

In recent years, there has been a proliferation of Edge Intelligence (EI) services, especially within Internet of Vehicles (IoV) scenarios, accompanied by a growing demand for multi-modal content generation. In response, Generative Artificial Intelligence (GAI) has emerged as a promising solution, equipping EI to produce diverse Artificial Intelligence-Generated Content (AIGC) for ubiquitous edge services. However, existing cloud-based GAI capabilities, which are mostly provided via the web and the Internet, introduce unacceptable latency overhead and heightened security risks for vehicular services. To address the above shortcomings and the lack of endogenous mechanisms for applying GAI to IoV scenarios, in this paper, we propose a layered vehicular GAI framework that seamlessly integrates GAI and EI. Within this framework, we devise a distributed collaborative inference mechanism between Road-Side Units (RSUs) and vehicles. Furthermore, we formulate the shared and local inference splitting problem, a pivotal challenge influencing both GAI service latency and content-generation capability. To tackle this issue, we introduce a backward induction-based algorithm, which enables the system can make splitting decisions using a simple threshold-based policy. Simulation results underscore the remarkable performance of the proposed system and vehicular collaborative inference mechanism, promising to facilitate diverse content generation within vehicular networks.

关键词： Cloud computing Inference mechanisms Generative AI Simulation Collaboration Inference algorithms Security

来源：评论

学校读者我要写书评

暂无评论

Large Scale Model Enabled Semantic Communications Based on Robust Knowledge Distillation

Large Scale Model Enabled Semantic Communications Based on R...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Kuiyuan Ding Fangfang Liu Yang Yang Mingzhe Chen Caili Guo Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China Department of Electrical and Computer Engineering and the Institute for Data Science and Computing University of Miami Coral Gables FL USA Beijing Laboratory of Advanced Information Networks School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

Large scale artificial intelligence (AI) models possess excellent capabilities in semantic representation and understanding, making them particularly well-suited for semantic encoding and decoding. However, the substantial scale of these AI models imposes unacceptable computational resources and communication delays. To address this issue, we propose a semantic communication scheme based on robust knowledge distillation (RKD-SC) for large scale model enabled semantic communications. In the considered system, a transmitter extracts the features of the source image for robust transmission and accurate image classification at the receiver. To effectively utilize the superior capability of large scale model while make the cost affordable, we first transfer knowledge from a large scale model to a smaller scale model to serve as the semantic encoder. Then, to enhance the robustness of the system against channel noise, we propose a channel-aware autoencoder (CAA) based on the Transformer architecture. Experimental results show that the encoder of proposed RKD-SC system can achieve over 93.3% of the performance of a large scale model while compressing 96.67% number of parameters. Code: https://***/echojayne/RKD-SC.

关键词： Transmitters Computational modeling Noise Autoencoders Receivers Semantic communication Transformers Feature extraction Robustness Decoding

来源：评论

学校读者我要写书评

暂无评论

CoG-DQA: Chain-of-Guiding Learning with Large Language Models for Diagram Question Answering

CoG-DQA: Chain-of-Guiding Learning with Large Language Model...

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Shaowei Wang Lingling Zhang Longji Zhu Tao Qin Kim-Hui Yap Xinyu Zhang Jun Liu School of Computer Science and Technology Xi'an Jiaotong University China Key Laboratory of Intelligent Networks and Network Security Xi'an Jiaotong University Ministry of Education Xi'an Shaanxi China School of Electrical & Electronic Engineering Nanyang Technological University Singapore

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Diagram Question Answering (DQA) is a challenging task, requiring models to answer natural language questions based on visual diagram contexts. It serves as a crucial basis for academic tutoring, technical support, and more practical applications. DQA poses significant challenges, such as the demand for domain-specific knowledge and the scarcity of annotated data, which restrict the applicability of large-scale deep models. Previous approaches have explored external knowledge integration through pretraining, but these methods are costly and can be limited by domain disparities. While Large Language Models (LLMs) show promise in question-answering, there is still a gap in how to cooperate and interact with the diagram parsing process. In this paper, we introduce the Chain-of-Guiding Learning Model for Diagram Question Answering (CoG-DQA), a novel framework that effectively addresses DQA challenges. CoG-DQA leverages LLMs to guide diagram parsing tools (DPTs) through the guiding chains, enhancing the precision of diagram parsing while introducing rich background knowledge. Our experimental findings reveal that CoG-DQA surpasses all comparison models in various DQA scenarios, achieving an average accuracy enhancement exceeding 5% and peaking at 11% across four datasets. These results underscore CoG-DQA's capacity to advance the field of visual question answering and promote the integration of LLMs into specialized domains.

关键词： Bridges Visualization computer vision Large language models Computational modeling Natural languages Question answering (information retrieval)

来源：评论

学校读者我要写书评

暂无评论

RedAgent: Red Teaming Large Language Models with Context-aware Autonomous Language Agent

arXiv

引用

arXiv 2024年

作者： Xu, Huiyu Zhang, Wenhui Wang, Zhibo Xiao, Feng Zheng, Rui Feng, Yunhe Ba, Zhongjie Ren, Kui The State Key Laboratory of Blockchain and Data Security Zhejiang University China School of Cyber Science and Technology Zhejiang University China Palo Alto Networks United States Department of Computer Science and Engineering University of North Texas United States

Recently, advanced Large Language Models (LLMs) such as GPT-4 have been integrated into many real-world applications like Code Copilot. These applications have significantly expanded the attack surface of LLMs, exposing them to a variety of threats. Among them, jailbreak attacks that induce toxic responses through carefully constructed jailbreak prompts have raised critical safety concerns. To effectively identify these threats, a growing number of red team approaches simulate potential adversarial scenarios by crafting jailbreak prompts to test the target LLM. However, existing red teaming methods do not consider the unique vulnerabilities of LLM in different scenarios (e.g., code-related tasks), making it difficult to adjust the jailbreak prompts to find context-specific vulnerabilities, thereby lacking efficiency. Meanwhile, these methods are limited to refining handcrafted jailbreak templates using a few mutation operations (such as synonym replacement), lacking the automation and scalability to continuously adapt to different scenarios. To enable context-aware and efficient red teaming, we abstract and model existing attacks into a coherent concept called "jailbreak strategy" and propose a multi-agent LLM system named RedAgent that leverages these strategies to generate context-aware jailbreak prompts. By self-reflecting on contextual feedback and red teaming trials in an additional memory buffer, RedAgent continuously learns how to leverage these strategies to achieve more effective jailbreaks in specific contexts. Extensive experiments demonstrate that our system can jailbreak most black-box LLMs within just five queries, improving the efficiency of existing red teaming methods by two times. Additionally, RedAgent can jailbreak customized LLM applications more efficiently. By generating context-aware jailbreak prompts towards trending applications on GPTs, we discover 60 severe vulnerabilities of these real-world applications with only two queries per vulnera

关键词： Autonomous agents

来源：评论

学校读者我要写书评

暂无评论

Covert Communication Based on Non-Ideal Detection of Overt Channels

Covert Communication Based on Non-Ideal Detection of Overt C...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Xinyu Zhao Danyang Wang Zan Li Lei Qian Chenxi Li Ning Zhang State Key Laboratory of Integrated Service Networks Xidian University Xi'an China Northwest Air Traffic Management Bureau Xi'an China Department of Electrical and Computer Engineering University of Windsor Windsor ON Canada

In this paper, we study a covert communication strategy based on non-ideal detection on overt channels in Internet of Things (IoT) networks, where IoT device utilizes existing overt channels as spectrum masks to achieve covert communication. We consider the non-ideal detection of IoT device on overt channels. At the same time, to improve covert transmission rate, we use improper Gaussian signaling (IGS) at IoT device. We first provide detection errors of IoT device, and then analyze the non-ideal transmission rate using IGS and minimum error detection probability of the warden. Next, we jointly optimize the transmit power of IoT device and circularity coefficient of IGS to maximize the covert rate while meeting the quality of service (QoS) constraint of the overt channel and the covertness constraint. The final simulations demonstrate that considering the non-ideal detection of IoT device and utilizing IGS can effectively improve the system transmission rate.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Knowledge Representation-Actuated Based Spatio-Temporal Graph Neural Network Traffic Flow Prediction

Knowledge Representation-Actuated Based Spatio-Temporal Grap...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Yihan Liu Nianwen Ning Ning Lu Yi Zhou School of Artificial Intelligence Henan University Zhengzhou China International Joint Research Laboratory for Cooperative Vehicular Networks of Henan Zhengzhou China Department of Electrical and Computer Engineering Queen's University Canada

In the task of traffic flow forecasting, various external factors need to be considered to interfere with the flow, such as weather conditions, traffic accidents, emergency events, and Points of Interest (POIs). While capturing the spatio-temporal dependencies, it is essential to effectively capture the external factors. However, existing studies cannot effectively cascade the information contained in these external factors to traffic features, and lack the co-capture of spatio-temporal features. To address these challenges, we present a Knowledge Representation learning-actuated Spatio-Temporal Graph Neural Network (KR-STGNN) for traffic flow prediction. The Gated Feature Fusion Module (GFFM) is utilized to combine the knowledge embedding with the traffic features, and the traffic features are updated adaptively and dynamically according to the importance of external factors. To conduct the co-capture of spatio-temporal dependencies, we subsequently propose a spatio-temporal feature synchronous capture module combining dilation causal convolution with GRU. Experimental results on a real-world traffic dataset demonstrate that KR-STGNN has superior forecasting performances with different prediction steps, especially for short-term prediction.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust Visual Question Answering: Datasets, Methods, and Future Challenges

arXiv

引用

arXiv 2023年

作者： Ma, Jie Wang, Pinghui Kong, Dechen Wang, Zewei Liu, Jun Pei, Hongbin Zhao, Junzhou Di, Jia The Ministry of Education of Key Laboratory for Intelligent Networks and Network Security School of Cyber Science and Engineering Xi’an Jiaotong University Shaanxi Xi’an710049 China The Ministry of Education of Key Laboratory for Intelligent Networks and Network Security School of Automation Science and Engineering Xi’an Jiaotong University Shaanxi Xi’an710049 China The Shannxi Provincial Key Laboratory of Big Data Knowledge Engineering School of Computer Science and Technology Xi’an Jiaotong University Shaanxi Xi’an710049 China

Visual question answering requires a system to provide an accurate natural language answer given an image and a natural language question. However, it is widely recognized that previous generic VQA methods often tend to memorize biases present in the training data rather than learning proper behaviors, such as grounding images before predicting answers. Therefore, these methods usually achieve high in-distribution but poor out-of-distribution performance. In recent years, various datasets and debiasing methods have been proposed to evaluate and enhance the VQA robustness, respectively. This paper provides the first comprehensive survey focused on this emerging fashion. Specifically, we first provide an overview of the development process of datasets from in-distribution and out-of-distribution perspectives. Then, we examine the evaluation metrics employed by these datasets. Thirdly, we propose a typology that presents the development process, similarities and differences, robustness comparison, and technical features of existing debiasing methods. Furthermore, we analyze and discuss the robustness of representative vision-and-language pre-training models on VQA. Finally, through a thorough review of the available literature and experimental analysis, we discuss the key areas for future research from various viewpoints. Copyright © 2023, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

FlowMur: A Stealthy and Practical Audio Backdoor Attack with Limited Knowledge

arXiv

引用

arXiv 2023年

作者： Lan, Jiahe Wang, Jie Yan, Baochen Yan, Zheng Bertino, Elisa The State Key Laboratory on Integrated Services Networks School of Cyber Engineering Xidian University China Hangzhou Institute of Technology Xidian University China Department of Computer Science Purdue University United States

Speech recognition systems driven by Deep Neural networks (DNNs) have revolutionized human-computer interaction through voice interfaces, which significantly facilitate our daily lives. However, the growing popularity of these systems also raises special concerns on their security, particularly regarding backdoor attacks. A backdoor attack inserts one or more hidden backdoors into a DNN model during its training process, such that it does not affect the model's performance on benign inputs, but forces the model to produce an adversary-desired output if a specific trigger is present in the model input. Despite the initial success of current audio backdoor attacks, they suffer from the following limitations: (i) Most of them require sufficient knowledge, which limits their widespread adoption. (ii) They are not stealthy enough, thus easy to be detected by humans. (iii) Most of them cannot attack live speech, reducing their practicality. To address these problems, in this paper, we propose FlowMur, a stealthy and practical audio backdoor attack that can be launched with limited knowledge. FlowMur constructs an auxiliary dataset and a surrogate model to augment adversary knowledge. To achieve dynamicity, it formulates trigger generation as an optimization problem and optimizes the trigger over different attachment positions. To enhance stealthiness, we propose an adaptive data poisoning method according to Signal-to-Noise Ratio (SNR). Furthermore, ambient noise is incorporated into the process of trigger generation and data poisoning to make FlowMur robust to ambient noise and improve its practicality. Extensive experiments conducted on two datasets demonstrate that FlowMur achieves high attack performance in both digital and physical settings while remaining resilient to state-ofthe- art defenses. In particular, a human study confirms that triggers generated by FlowMur are not easily detected by participants. The source code of FlowMur is publicly available at https://

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Few-shot Open Relation Extraction with Gaussian Prototype and Adaptive Margin

arXiv

引用

arXiv 2024年

作者： Guo, Tianlin Zhang, Lingling Wang, Jiaxin Lei, Yunkuo Li, Yifei Wang, Haofen Liu, Jun School of Computer Science and Technology Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi'an Jiaotong University Xi’an710049 China School of Computer Science and Technology Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi'an Jiaotong University Xi’an710049 China College of Design and Innovation Tongji University Shanghai200092 China

Few-shot relation extraction with none-of-the-above (FsRE with NOTA) aims at predicting labels in few-shot scenarios with unknown classes. FsRE with NOTA is more challenging than the conventional few-shot relation extraction task, since the boundaries of unknown classes are complex and difficult to learn. Meta-learning based methods, especially prototype-based methods, are the mainstream solutions to this task. They obtain the classification boundary by learning the sample distribution of each class. However, their performance is limited because few-shot overfitting and NOTA boundary confusion lead to misclassification between known and unknown classes. To this end, we propose a novel framework based on Gaussian prototype and adaptive margin named GPAM for FsRE with NOTA, which includes three modules, semi-factual representation, GMM-prototype metric learning and decision boundary learning. The first two modules obtain better representations to solve the few-shot problem through debiased information enhancement and Gaussian space distance measurement. The third module learns more accurate classification boundaries and prototypes through adaptive margin and negative sampling. In the training procedure of GPAM, we use contrastive learning loss to comprehensively consider the effects of range and margin on the classification of known and unknown classes to ensure the model's stability and robustness. Sufficient experiments and ablations on the FewRel dataset show that GPAM surpasses previous prototype methods and achieves state-of-the-art performance. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A Method for Person Re-Identification and Trajectory Tracking in Cross-Device Edge Environments

A Method for Person Re-Identification and Trajectory Trackin...

引用

International Conference on Mobile Ad-hoc and Sensor networks, MSN

作者： Jian An Yunxiang Zhao Feifei Wang Xin He Xiaolin Gui School of Computer Science and Technology Shaanxi Provincial Key Laboratory of Computer Networks Xi’an Jiaotong University Xi’an China Henan Jiaoyuan Engineering Technology Group Co. LTD Henan China School of Software Henan University Henan China

ISBN: (数字)9798350358261

ISBN: (纸本)9798350358278

Addressing expanded surveillance scopes, cross-device person re-identification emerges as a crucial issue. Traditional approaches to re-identification and tracking confront challenges in feature matching challenges, and fail to harness the potential of edge computing environments, which offer viable solutions to bandwidth and latency constraints. The paper introduces a novel, edge computing and shared cache-based method for person re-identification and tracking. In this framework, edge nodes exploit deep learning method to execute image recognition for person re-identification and trajectory tracking. Edge nodes are deployed within a tri-layered architecture, encompassing Hierarchical Edge Cloud (HEC), Mobile Edge Computing (MEC), and Central Cloud (CC). Emphasizing part feature significance for person re-identification, the approach employs a refined Part-based Convolutional Baseline model guided by the triplet loss function. Experimental outcomes validate the high re-identification precision and efficacious load balancing capability of this method in person re-identification and tracking operations. Achieving a harmonized trade-off between task accuracy and response time, the proposed framework offers a robust solution for person re-identification and tracking within edge computing contexts.

关键词： Accuracy Trajectory tracking Image edge detection Computational modeling computer architecture Trajectory Time factors

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：