检索结果-内蒙古大学图书馆

23rd USENIX Conference on File and Storage Technologies, FAST 2025

作者： Chen, Weijian He, Shuibing Qu, Haoyang Zhang, Ruidong Yang, Siling Chen, Ping Zheng, Yi Huai, Baoxing Chen, Gang The State Key Laboratory of Blockchain and Data Security Zhejiang University China Institute of Blockchain and Data Security China Zhejiang Key Laboratory of Big Data Intelligent Computing China Huawei Cloud China

ISBN: (纸本)9781939133458

Modern advanced large language model (LLM) applications often prepend long contexts before user queries to improve model output quality. These contexts frequently repeat, either partially or fully, across multiple queries. Existing systems typically store and reuse the keys and values of these contexts (referred to as prefix KVs) to reduce redundant computation and time to first token (TTFT). When prefix KVs need to be stored on disks due to insufficient CPU memory, reusing them does not always reduce TTFT, as disk I/O latency is high. In this paper, we propose IMPRESS, an importance-informed multi-tier prefix KV storage system to reduce I/O delay for LLM inference by only loading important prefix KVs. IMPRESS first leverages the insight that there is significant similarity in important token index sets across attention heads and introduces an I/O-efficient important KV identification algorithm. It then optimizes prefix KV storage and caching through importance-informed KV management, reducing TTFT during model inference. Our experimental results show that IMPRESS can reduce TTFT by up to 2.8× compared to state-of-the-art systems, while maintaining comparable inference accuracy. © 2025 FAST. All Rights Reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Weak Multi-Label data Stream Classification Under Distribution Changes in Labels

引用

IEEE Transactions on big data 2025年第3期11卷 1369-1380页

作者： Zou, Yizhang Hu, Xuegang Li, Peipei Hu, Jun Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data Ministry of Education School of Computer Science and Information Engineering Hefei230002 China National University of Singapore School of Computing 119077 Singapore

Multi-label stream classification aims to address the challenge of dynamically assigning multiple labels to sequentially arrived instances. In real situations, only partial labels of instances can be observed due to the expensive human annotations, and the problem of label distribution changes arises from multiple labels in a streaming mode, but few existing works jointly consider such challenges. Motivated by this, we propose the problem of weak multi-label stream classification (WMSC) and an online classification algorithm robust to weak labels. Specifically, we incrementally update the margin-based model using information from both the past model and the current incoming instance with partially observed labels. To increase the robustness to weak labels, we first adjust the classification margin of negative labels using the label causality matrix, which is constructed by the conditional probability of label pairs. Second, we introduce the label prototype matrix to regulate the margin by controlling the weighting parameter of the slack term. Additionally, to handle the potential distribution changes in labels, we utilize the instance-specific threshold via online thresholding to perform binary classification, which is formulated as a regression problem. Finally, theoretical analysis and empirical experimental results are presented to demonstrate the effectiveness of WMSC in classifying unobserved streaming instances. © 2015 IEEE.

关键词： Labeled data

来源：评论

学校读者我要写书评

暂无评论

Attribute-Based Access Control Encryption

引用

IEEE Transactions on Dependable and Secure computing 2025年第3期22卷 2227-2242页

作者： Wang, Xiuhua Yu, Mengyang Wang, Yaohui Pi, Yinjia Xu, Peng Wang, Shuai Jin, Hai Han, Min Huazhong University of Science and Technology Hubei Key Laboratory of Distributed System Security School of Cyber Science and Engineering Wuhan430074 China State Key Laboratory of Cryptology Beijing100878 China Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Wuhan430074 China University of Electronic Science and Technology of China National Key Laboratory of Wireless Communications Chengdu611731 China Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Wuhan430074 China

The burgeoning complexity of communication necessitates a high demand for security. Access control encryption is a promising primitive to meet the security demand but the bulk of its constructions rely on formulating the access control policy with identities. Attribute-based access control policy in attribute-based encryption (ABE) is known to be more expressive without relying on enumerating identities. We propose a generic framework to build attribute-based access control encryption from ciphertext-policy ABE. Our instantiations prioritize different emphases on expressiveness and efficiency. The first instantiation supports multi-valued AND-gate access control structures, while the second supports the linear-secret-sharing access structure. Both are prototyped with efficiency validated empirically. © 2024 IEEE.

关键词： Ciphertext

来源：评论

学校读者我要写书评

暂无评论

A numerical magnitude aware multi-channel hierarchical encoding network for math word problem solving

引用

Neural computing and Applications 2025年第3期37卷 1651-1672页

作者： Xu, Junjie Chen, Yuzhong Xiao, Lingsheng Liao, Hongmiao Zhong, Jiayuan Dong, Chen College of Computer and Data Science Fuzhou University Fujian Province Fuzhou350108 China Engineering Research Center of Big Data Intelligence Ministry of Education Fuzhou China Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fujian Province Fuzhou350108 China

Math word problem (MWP) represents a critical research area within reading comprehension, where accurate comprehension of math problem text is crucial for generating math expressions. However, current approaches still grapple with unresolved challenges in grasping the sensitivity of math problem text and delineating distinct roles across various clause types, and enhancing numerical representation. To address these challenges, this paper proposes a Numerical Magnitude Aware Multi-Channel Hierarchical Encoding Network (NMA-MHEA) for math expression generation. Firstly, NMA-MHEA implements a multi-channel hierarchical context encoding module to learn context representations at three different channels: intra-clause channel, inter-clause channel, and context-question interaction channel. NMA-MHEA constructs hierarchical constituent-dependency graphs for different levels of sentences and employs a Hierarchical Graph Attention Neural Network (HGAT) to learn syntactic and semantic information within these graphs at the intra-clause and inter-clause channels. NMA-MHEA then refines context clauses using question information at the context-question interaction channel. Secondly, NMA-MHEA designs a number encoding module to enhance the relative magnitude information among numerical values and type information of numerical values. Experimental results on two public benchmark datasets demonstrate that NMA-MHEA outperforms other state-of-the-art models. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

An Evolutionary Multitasking Algorithm for Efficient Multiobjective Recommendations

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第3期6卷 518-532页

作者： Tian, Ye Ji, Luke Hu, Yiwei Ma, Haiping Wu, Le Zhang, Xingyi Anhui University Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Hefei230601 China Anhui University Institutes of Physical Science and Information Technology Hefei230601 China Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data Hefei230029 China

Represented by evolutionary algorithms and swarm intelligence algorithms, nature-inspired metaheuristics have been successfully applied to recommender systems and amply demonstrated effectiveness, in particular, for multiobjective recommendation. Owing to the population-based search paradigm, these algorithms can produce a number of recommendation lists, making diverse tradeoffs between multiple metrics and meeting the requirements of accuracy, novelty, diversity, and other user preferences. However, these algorithms are criticized for the low efficiency of the optimization process, especially when the number of users is large. To address this issue, this article proposes an evolutionary multitasking-based recommendation method, where each task corresponds to a user and all the tasks are optimized simultaneously, thus highly improving the efficiency of recommendation. To enhance the convergence speed, all the users are divided into multiple populations according to the similarity between their preferences, where each population evolves with internal knowledge transfer between users, and all the populations evolve with external knowledge transfer between populations. Experimental results on various datasets verify that the proposed method can better balance between multiple metrics than classical and deep neural network-based recommendation methods and exhibits significantly higher efficiency than evolutionary multiobjective optimization-based recommendation methods. © 2024 IEEE. All rights reserved.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

T³Planner: Multi-Phase Planning Across Structure-Constrained Optical, IP, and Routing Topologies

引用

IEEE Journal on Selected Areas in Communications 2025年第5期43卷 1823-1839页

作者： Hao, Yijun Yang, Shusen Li, Fang Zhang, Yifan Zhao, Cong Ren, Xuebin Zhao, Peng Xu, Chenren Wang, Shibo Xi’an Jiaotong University National Engineering Laboratory for Big Data Analytics Xi’an710049 China Xi’an Jiaotong University National Engineering Laboratory for Big Data Analytics Ministry of Education Key Laboratory for Intelligent Networks and Network Security Xi’an710049 China Peking University School of Computer Science Beijing100871 China The Chinese University of Hong Kong Department of Computer Science and Engineering Hong Kong National Engineering Laboratory for Big Data Analytics Xi'an Jiaotong University Xi'an710049 China

Network topology planning is an essential multi-phase process to build and jointly optimize the multi-layer network topologies in wide-area networks (WANs). Most existing practices target single-phase/layer planning, and are incapable of satisfying all rigorous topological structure constraints (e.g., dual-homing rings) defined by network standards and operators, especially in large-scale networks. These significantly limit their usability and performance in production networks. We consider a general topology planning problem with typical structure constraints over three essential phases (greenfield, reconfiguration, and site expansion) and topological layers (optical, IP, and routing topologies). We present, T3Planner, a novel practical solver to this problem in production. Specifically, we develop a structure-driven encoder based on graph neural network (GNN) for concise structure encoding, and design a new learning framework with optical-centric layer compression/reconstruction and rule-aided reinforcement learning (RL) for fast convergence and high performance. Extensive experiments on nine real topologies demonstrate that T3Planner scales to large optical networks with hundreds of sites, saves 46.6% cost, and supports 3.12x more demand when compared to related existing approaches. © 1983-2012 IEEE.

关键词： Network topology

来源：评论

学校读者我要写书评

暂无评论

Face Forgery Detection Based on Fine-Grained Clues and Noise Inconsistency

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第1期6卷 144-158页

作者： Zhang, Dengyong He, Ruiyi Liao, Xin Li, Feng Chen, Jiaxin Yang, Gaobo Changsha University of Science and Technology Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation the School of Computer and Communication Engineering Changsha410114 China Hunan University Changsha410082 China

Deepfake detection has gained increasing research attention in media forensics, and a variety of works have been produced. However, subtle artifacts might be eliminated by compression, and the convolutional neural networks (CNNs)-based detectors are invalidated for fake face images with compression. In this work, we propose a two-stream network for deepfake detection. We observed that high-frequency noise features and spatial features are inherently complementary to each other. Thus, both spatial features and high-frequency noise features are exploited for face forgery detection. Specifically, we design a double-frequency transformer module (DFTM) to guide the learning of spatial features from local artifact regions. To effectively fuse spatial features and high-frequency noise features, a dual-domain attention fusion module (DDAFM) is designed. We also introduce a local relationship constraint loss, which requires only image-level labels, for model training. We evaluate the proposed approach on five large-scale benchmark datasets, and extensive experimental results demonstrate the proposed approach outperforms most SOTA works. © 2024 IEEE.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

DMC-Watermark: A backdoor richer watermark for dual identity verification by dynamic mask covering

引用

Applied Intelligence 2025年第8期55卷 1-15页

作者： Zhu, Yujia Wang, Ruoxi Xia, Daoxun School of Big Data and Computer Science Guizhou Normal University Guiyang China Guizhou Key Laboratory of Advanced Computing Guizhou Normal University Guiyang China Engineering Laboratory for Applied Technology of Big Data in Education Guizhou Normal University Guiyang China

With the increasing use of neural networks, the importance of copyright protection for these models has gained significant attention. Backdoor watermarking is one of the key methods for protecting copyright. However, on the one hand, most existing backdoor watermarks are triggered by visual images, making them easily detectable, and therefore vulnerable to various attacks. On the other hand, it is difficult for these methods to carry information related to the creator’s identity which can easily lead to fraudulent claims of ownership. These factors contribute to the vulnerability and limitations of backdoor watermarking. In this paper, we propose DMC-Watermark, a backdoor richer watermarking method that uses dynamic mask-covered image structures as triggers. Leveraging the semantic preservation of image structure in transformation attacks, we select image structure as triggers. Furthermore, we convert the author-related information into an array of color information and apply it as a mask to the extracted image structures, allowing it to serve as a second layer of verification during the validation phase to resist fraudulent claims of ownership. The final trigger pattern, embedded with author-related image structures, is applied to the selected images in the trigger set, generating a final trigger set that is trained together with clean samples to produce a protected model. The experiments show that the proposed DMC-Watermark performs well in terms of fidelity, invisibility, undetectability, functionality, dual verification and robustness on three different datasets and four representative DNNs, and it has wide applicability and excellent results in high-resolution images. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Image watermarking

来源：评论

学校读者我要写书评

暂无评论

Gupacker: Generalized Unpacking Framework for Android Malware

引用

IEEE Transactions on Information Forensics and Security 2025年 20卷 4338-4352页

作者： Zheng, Tao Hou, Qiyu Chen, Xingshu Ren, Hao Li, Meng Li, Hongwei Shen, Changxiang Sichuan University School of Cyber Science and Engineering Chengdu610065 China Sichuan University School of Cyber Science and Engineering Cyber Science Research Institute Key Laboratory of Data Protection and Intelligent Management Ministry of Education Chengdu610065 China Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Intelligent Interconnected Systems Laboratory of Anhui Province School of Computer Science and Information Engineering Hefei230002 China University of Padua Department of Mathematics HIT Center Padua35131 Italy University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu611731 China Sichuan University Cyber Science Research Institute Key Laboratory of Data Protection and Intelligent Management Ministry of Education Chengdu610065 China

Android malware authors often use packers to evade analysis. Although many unpacking tools have been proposed, they face two significant challenges: 1) They are easily impeded by anti-analysis techniques employed by packers, preventing efficient collection of hidden Dex data. 2) They are typically designed to unpack a specific packer and cannot handle malware packed with mixed packers. Consequently, many packed malware samples evade detection. To bridge this gap, we propose Gupacker , a novel generalized unpacking framework. Gupacker offers a generic solution for first-generation holistic packer by customizing the Android system source code. It identifies the type of packer and selects an appropriate unpacking function, constructs a deeper active call chain to achieve generic unpacking of second-generation function extraction packers, and uses JNI function and instruction monitoring to handle third-generation virtual obfuscation packer. On this basis, we counteract a diverse array of anti-analysis techniques. We conduct extensive experiments on 5K packed Android malware samples, comparing Gupacker with 2 commercial and 4 state-of-the-art academic unpacking tools. The results demonstrate that Gupacker significantly improves the efficiency of Android malware unpacking with acceptable system overhead. We analyze real packed applications based on Gupacker and found several are second-packed by attackers, including WPS for Android, with tens of millions of users. We receive and responsibly report 13 0day vulnerabilities and also assist in the remediation of all vulnerabilities. © 2005-2012 IEEE.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

LeapGNN: Accelerating Distributed GNN Training Leveraging Feature-Centric Model Migration 23

LeapGNN: Accelerating Distributed GNN Training Leveraging Fe...

引用

23rd USENIX Conference on File and Storage Technologies, FAST 2025

作者： Chen, Weijian He, Shuibing Qu, Haoyang Zhang, Xuechen The State Key Laboratory of Blockchain and Data Security Zhejiang University China Zhejiang Lab China Institute of Blockchain and Data Security China Zhejiang Key Laboratory of Big Data Intelligent Computing China Washington State University Vancouver United States

ISBN: (纸本)9781939133458

Distributed training of graph neural networks (GNNs) has become a crucial technique for processing large graphs. Prevalent GNN frameworks are model-centric, necessitating the transfer of massive graph vertex features to GNN models, which leads to a significant communication bottleneck. Recognizing that the model size is often significantly smaller than the feature size, we propose LeapGNN, a feature-centric framework that reverses this paradigm by bringing GNN models to vertex features. To make it truly effective, we first propose a micrograph-based training strategy that leverages a refined structure to enhance locality, combined with the model migration technique, to minimize remote feature retrieval. Then, we devise a feature pre-gathering approach that merges multiple fetch operations into a single one to eliminate redundant feature transmissions. Finally, we employ a micrograph-based merging method that adjusts the number of micrographs for each worker to minimize kernel switches and synchronization overhead. Our experimental results demonstrate that LeapGNN achieves a performance speedup of up to 4.2× compared to the state-of-the-art method, namely P3 © 2025 FAST. All Rights Reserved.

关键词： Graph neural networks

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

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

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：