检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Wu, Jialong Wang, Zhenglin Zhang, Linhai Lai, Yilong He, Yulan Zhou, Deyu School of Computer Science and Engineering Key Laboratory of Computer Network and Information Integration Ministry of Education Southeast University China Department of Informatics King’s College London United Kingdom The Alan Turing Institute United Kingdom

Key-Value (KV) cache has become a bottleneck of LLMs for long-context generation. Despite the numerous efforts in this area, the optimization for the decoding phase is generally ignored. However, we believe such optimization is crucial, especially for long-output generation tasks based on the following two observations: (i) Excessive compression during the prefill phase which requires specific full context, impairs the comprehension of the reasoning task;(ii) Deviation of heavy hitters1 occurs in the reasoning tasks with long outputs. Therefore, SCOPE, a simple yet efficient framework that separately performs KV cache optimization during the prefill and decoding phases, is introduced. Specifically, the KV cache during the prefill phase is preserved to maintain the essential information, while a novel strategy based on sliding is proposed to select essential heavy hitters for the decoding phase. Memory usage and memory transfer are further optimized using adaptive and discontinuous strategies. Extensive experiments on LONGGENBENCH show the effectiveness and generalization of SCOPE and its compatibility as a plug-in to other prefill-only KV compression methods. 2 Copyright © 2024, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Causal Prompting: Debiasing Large Language Model Prompting based on Front-Door Adjustment

arXiv

引用

arXiv 2024年

作者： Zhang, Congzhi Zhang, Linhai Wu, Jialong He, Yulan Zhou, Deyu School of Computer Science and Engineering Key Laboratory of Computer Network and Information Integration Ministry of Education Southeast University China Department of Informatics King’s College London United Kingdom The Alan Turing Institute United Kingdom

Despite the notable advancements of existing prompting methods, such as In-Context Learning and Chain-of-Thought for Large Language Models (LLMs), they still face challenges related to various biases. Traditional debiasing methods primarily focus on the model training stage, including approaches based on data augmentation and reweighting, yet they struggle with the complex biases inherent in LLMs. To address such limitations, the causal relationship behind the prompting methods is uncovered using a structural causal model, and a novel causal prompting method based on front-door adjustment is proposed to effectively mitigate LLMs biases. In specific, causal intervention is achieved by designing the prompts without accessing the parameters and logits of LLMs. The chain-of-thought generated by LLM is employed as the mediator variable and the causal effect between input prompts and output answers is calculated through front-door adjustment to mitigate model biases. Moreover, to accurately represent the chain-of-thoughts and estimate the causal effects, contrastive learning is used to fine-tune the encoder of chain-of-thought by aligning its space with that of the LLM. Experimental results show that the proposed causal prompting approach achieves excellent performance across seven natural language processing datasets on both open-source and closed-source LLMs. 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

The Power of Bamboo: On the Post-Compromise Security for Searchable Symmetric Encryption 30

The Power of Bamboo: On the Post-Compromise Security for Sea...

引用

30th Annual Network and Distributed system Security Symposium, NDSS 2023

作者： Chen, Tianyang Xu, Peng Picek, Stjepan Luo, Bo Susilo, Willy Jin, Hai Liang, Kaitai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab China Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering China Cluster and Grid Computing Lab School of Computer Science and Technology China Huazhong University of Science and Technology Wuhan430074 China Digital Security Group Radboud University Nijmegen Netherlands Department of EECS Institute of Information Sciences The University of Kansas LawrenceKS United States Institute of Cybersecurity and Cryptology School of Computing and Information Technology University of Wollongong WollongongNSW2522 Australia Faculty of Electrical Engineering Mathematics and Computer Science Delft University of Technology Delft2628 CD Netherlands

ISBN: (纸本)1891562835

Dynamic searchable symmetric encryption (DSSE) enables users to delegate the keyword search over dynamically updated encrypted databases to an honest-but-curious server without losing keyword privacy. This paper studies a new and practical security risk to DSSE, namely, secret key compromise (e.g., a user’s secret key is leaked or stolen), which threatens all the security guarantees offered by existing DSSE schemes. To address this open problem, we introduce the notion of searchable encryption with key-update (SEKU) that provides users with the option of non-interactive key updates. We further define the notion of post-compromise secure with respect to leakage functions to study whether DSSE schemes can still provide data security after the client’s secret key is compromised. We demonstrate that post-compromise security is achievable with a proposed protocol called "Bamboo". Interestingly, the leakage functions of Bamboo satisfy the requirements for both forward and backward security. We conduct a performance evaluation of Bamboo using a real-world dataset and compare its runtime efficiency with the existing forward-and-backward secure DSSE schemes. The result shows that Bamboo provides strong security with better or comparable performance. © 2023 30th Annual Network and Distributed system Security Symposium, NDSS 2023. All Rights Reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

BAT: Mining Binary-API Topic for Multi-service Application Development

BAT: Mining Binary-API Topic for Multi-service Application D...

引用

International Conference on computer Supported Cooperative Work in Design

作者： Pengfei He Linlin Liu Dianlong You Limin Shen Zhen Chen School of Information Science and Engineering Yanshan University Qinhuangdao China National Science Libraries Chinese Academy of Sciences Beijing China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao China

With the growing popularity of API-driven multiservice application (mashup) development, the burgeoning web APIs have left developers drowning in the sea of web API selections. Matching developers with the most appropriate APIs is the key to improving user satisfaction and promoting more popular web applications. As a result, more and more researchers pay attention to web API recommender systems based on collaborative filtering. However, employing collaborative filtering to recommend APIs is challenging due to the severe sparsity of mashup and API interactions. To address this problem, we propose a probabilistic generative model, called the Binary-API Topic model (BAT), to parameterize mashups and APIs. Technically, BAT is equipped with a mechanism to extract binary-APIs and predict unknown pairwise interactions. To improve generality and capture more relevance from a limited number of interactions, we learn binary-API topics by directly modeling the generation of API co-occurrence patterns across the repository (all mashup collections from ***). The main advantage of BAT is that it preserves API co-occurrence patterns in model learning and exploits the rich global relevance. Finally, through extensive experiments, we demonstrate that BAT can achieve the highest performance on the sparse real-world data set.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Continual Reinforcement Learning for Digital Twin Synchronization Optimization

arXiv

引用

arXiv 2025年

作者： Tong, Haonan Chen, Mingzhe Zhao, Jun Hu, Ye Yang, Zhaohui Liu, Yuchen Yin, Changchuan The Beijing Key Laboratory of Network System Architecture and Convergence The Beijing Advanced Information Network Laboratory Beijing University of Posts and Telecommunications Beijing China Singapore The Department of Electrical and Computer Engineering Frost Institute for Data Science and Computing University of Miami Coral GablesFL33146 United States The Department of Industrial and Systems Engineering University of Miami Coral GablesFL33146 United States The College of Information Science and Electronic Engineering Zhejiang University Zhejiang Hangzhou310027 China The Department of Computer Science North Carolina State University RaleighNC27695 United States The Beijing Key Laboratory of Network System Architecture and Convergence China The Beijing Advanced Information Network Laboratory Beijing University of Posts and Telecommunications Beijing100876 China

This article investigates the adaptive resource allocation scheme for digital twin (DT) synchronization optimization over dynamic wireless networks. In our considered model, a base station (BS) continuously collects factory physical object state data from wireless devices to build a real-time virtual DT system for factory event analysis. Due to continuous data transmission, maintaining DT synchronization must use extensive wireless resources. To address this issue, a subset of devices is selected to transmit their sensing data, and resource block (RB) allocation is optimized. This problem is formulated as a constrained Markov process (CMDP) problem that minimizes the long-term mismatch between the physical and virtual systems. To solve this CMDP, we first transform the problem into a dual problem that refines RB constraint impacts on device scheduling strategies. We then propose a continual reinforcement learning (CRL) algorithm to solve the dual problem. The CRL algorithm learns a stable policy across historical experiences for quick adaptation to dynamics in physical states and network capacity. Simulation results show that the CRL can adapt quickly to network capacity changes and reduce normalized root mean square error (NRMSE) between physical and virtual states by up to 55.2%, using the same RB number as traditional methods. © 2025, CC BY-NC-SA.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-Preserving Talking Face Synthesis

arXiv

引用

arXiv 2024年

作者： Deng, Kaijun Zheng, Dezhi Xie, Jindong Wang, Jinbao Xie, Weicheng Shen, Linlin Song, Siyang Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China Department of Computer Science University of Exeter United Kingdom

Accurately synthesizing talking face videos and capturing fine facial features for individuals with long hair presents a significant challenge. To tackle these challenges in existing methods, we propose a decomposed per-embedding Gaussian fields (DEGSTalk), a 3D Gaussian Splatting (3DGS)-based talking face synthesis method for generating realistic talking faces with long hairs. Our DEGSTalk employs Deformable Pre-Embedding Gaussian Fields, which dynamically adjust pre-embedding Gaussian primitives using implicit expression coefficients. This enables precise capture of dynamic facial regions and subtle expressions. Additionally, we propose a Dynamic Hair-Preserving Portrait Rendering technique to enhance the realism of long hair motions in the synthesized videos. Results show that DEGSTalk achieves improved realism and synthesis quality compared to existing approaches, particularly in handling complex facial dynamics and hair preservation. Our code will be publicly available at https://***/CVI-SZU/DEGSTalk. Copyright © 2024, The Authors. All rights reserved.

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Short-term multi-step-ahead sector-based traffic flow prediction based on the attention-enhanced graph convolutional LSTM network (AGC-LSTM)

引用

Neural Computing and Applications 2024年 1-20页

作者： Zhang, Ying Xu, Shimin Zhang, Linghui Jiang, Weiwei Alam, Sameer Xue, Dabin College of General Aviation and Flight Nanjing University of Aeronautics and Astronautics Nanjing China State Key Laboratory of Air Traffic Management System Nanjing China College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China Air Traffic Management Research Institute School of Mechanical & ampAerospace Engineering Nanyang Technological University Singapore Singapore Department of Land Surveying and Geo-Informatics The Hong Kong Polytechnic University Hong Kong

Accurate sector-based air traffic flow predictions are essential for ensuring the safety and efficiency of the air traffic management (ATM) system. However, due to the inherent spatial and temporal dependencies of air traffic flow, it is still a challenging problem. To solve this problem, some methods are proposed considering the relationship between sectors, while the complicated spatiotemporal dynamics and interdependencies between traffic flow of route segments related to the sector are not taken into account. To address this challenge, the attention-enhanced graph convolutional long short-term memory network (AGC-LSTM) model is applied to improve the short-term sector-based traffic flow prediction, in which spatial structures of route segments related to the sector are considered for the first time. Specifically, the graph convolutional networks (GCN)-LSTM network model was employed to capture spatiotemporal dependencies of the flight data, and the attention mechanism is designed to concentrate on the informative features from key nodes at each layer of the AGC-LSTM model. The proposed model is evaluated through a case study of the typical enroute sector in the central–southern region of China. The prediction results show that MAE reduces by 14.4% compared to the best performing GCN-LSTM model among the other five models. Furthermore, the study involves comparative analyses to assess the influence of route segment range, input and output sequence lengths, and time granularities on prediction performance. This study helps air traffic managers predict flight situations more accurately and avoid implementing overly conservative or excessively aggressive flow management measures for the sectors. © The Author(s) 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Candidate Evaluation with Multimodal Data-Driven for Recruitment 27th

Candidate Evaluation with Multimodal Data-Driven for Recru...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Wu, Xing Liu, Kehong Wang, Jianjia Yao, Junfeng Deng, Bin Lv, Rongqi Song, Jun School of Computer Engineering and Science Shanghai University Shanghai200444 China Shanghai Institute for Advanced Communication and Data Science Shanghai University Shanghai200444 China School of AI and Advanced Computing XJTLU Entrepreneur College Xi’an Jiaotong-Liverpool University Suzhou215412 China CSSC Seago System Technology Co. Ltd. Shanghai200010 China Science and Technology on Near-Surface Detection Laboratory Jangsu Wuxi214035 China Department of Geography Hong Kong Baptist University 999077 Hong Kong

ISBN: (纸本)9783031781858

In the field of intelligent recruitment, automated resume matching and non-contact interviews have significantly improved the efficiency of companies in finding suitable candidates. This corresponds to the techniques of person-job matching and AI interviews. However, current person-job matching methods lack substantial data support, while AI interview methods struggle to integrate deep information from multimodal data and provide comprehensive evaluations of candidates’ responses. To address these challenges, we propose a multimodal data-driven person-job evaluation model, comprising two key stages: a person-job matching method based on graph attention and a multimodal AI interview method. Using a dual-perspective graph neural network approach, we accomplish the screening of candidates and positions. In the second stage, we conduct a comprehensive evaluation of candidates’ interview performance based on text, audio, and image modalities, providing a more objective, consistent, and efficient interview assessment method. Experimental results demonstrate that our person-job matching method surpasses current popular techniques and effectively transfers features to the next stage. In our multimodal AI interview method, we achieve accurate scoring of candidate responses, assessment of intonation stress levels, and inference of their Big Five personality traits, comprehensively evaluating candidates from multiple perspectives. This confirms the superiority and efficiency of our approach. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Streamflow Prediction in the Mekong River Basin Using Deep Neural Networks

引用

IEEE Access 2023年 11卷 97930-97943页

作者： Nguyen, Thi-Thu-Ha Vu, Duc-Quang Mai, Son T. Dang, Thanh Duc National Taiwan University Department of Bioenvironmental Systems Engineering Taipei10617 Taiwan National Central University Department of Computer Science and Information System Taoyuan320317 Taiwan Thai Nguyen University of Education Thai Nguyen250000 Viet Nam Queen's University Belfast School of Electronics Electrical Engineering and Computer Science BelfastBT7 1NN United Kingdom Institute for Computational Science and Artificial Intelligence Van Lang University Laboratory of Environmental Sciences and Climate Change Ho Chi Minh City70000 Viet Nam Van Lang University Faculty of Environment Ho Chi Minh City70000 Viet Nam

In recent years, the Mekong River Basin (MRB), one of the largest river basins in Southeast Asia, has experienced severe impacts from extreme droughts and floods. Streamflow forecasting has become crucial for effective risk management strategies in the region. However, this task presents significant challenges due to rapid climate changes and the presence of numerous newly constructed upstream dams, which disrupt the natural flow. In this paper, we develop multiple deep learning models (incl. Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), Long short-term Memory (LSTM), and Transformer) to predict streamflow with different lead time forecasts based on observed meteorological variables and climatic indices (i.e., discharge, water level, precipitation, and temperature) from 1979 to 2019. The results indicate that LSTM obtains high performance for streamflow prediction in both dry and wet seasons while Transformer is not recommended for long-term prediction, especially in the dry season. The proposed deep learning models capture well the fluctuation of river flow in the MRB during the period of high-dam development, especially LSTM (NSE ≥ 0.8). The models' performances are enhanced with the adding of temperature for short-term prediction while precipitation was the most sensitive variable for long-term one. Such proposed models are essential for government agencies to plan mitigation and adaptation strategies at different periods, which can range from days to years. © 2013 IEEE.

关键词： Floods

来源：评论

学校读者我要写书评

暂无评论

Enhancing Security in OpenID Connect Single Sign-On (SSO) systems: A Comprehensive Vulnerability Analysis and Mitigation Approach 4

Enhancing Security in OpenID Connect Single Sign-On (SSO) Sy...

引用

4th International Conference on Distributed Sensing and Intelligent systems, ICDSIS 2023

作者： Al-Shabi, Mohamed Bennour, Akram Al-Sarem, Mahamed Khalaf, Osama Ibrahim Sikder, Rajesh Algburi, Sameer Department of management Information system Taibah University Medina Saudi Arabia "LAMIS" Laboratory Echahid Chikh Larbi Tebessi University Tebessa Algeria College of Computer Science and Engineering Taibah University Medina Saudi Arabia Al-Nahrain University Jadriya Baghdad Iraq Information Technology University of the Cumberlands United States Al-Kitab University College of Engineering Techniques Kirkuk Iraq

ISBN: (纸本)9781839539954

Single Sign-On (SSO) systems have become popular for simplifying the login process by allowing users to authenticate with a single identity provider (IDP). However, the widespread use of these systems raises concerns about user privacy, as IDPs like Google or Facebook can gather extensive information about user behavior on the web. This poses a significant challenge for privacy-conscious users who wish to limit the disclosure of their web access behavior to third-party organizations. In this paper, we present a comprehensive and large-scale practical study focused on the OpenID Connect protocol, a widely used SSO protocol. Our analysis delves into the various stages of the protocol's operation, highlighting its security flaws and vulnerabilities. Additionally, we conduct a systematic examination of the privacy implications associated with user access to SSO systems. We provide a detailed account of how easily user information can be accessed, shedding light on the potential risks. © The Institution of engineering & Technology 2023.

关键词： Differential privacy

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：