检索结果-内蒙古大学图书馆

Not All Image Regions Matter: Masked Vector Quantization for Autoregressive Image Generation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Huang, Mengqi Mao, Zhendong Wang, Quan Zhang, Yongdong University of Science and Technology of China Hefei China MOE Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications Beijing China Institute of Artificial intelligence Hefei Comprehensive National Science Center Hefei China

Existing autoregressive models follow the two-stage generation paradigm that first learns a codebook in the latent space for image reconstruction and then completes the image generation autoregressively based on the learned codebook. However, existing codebook learning simply models all local region information of images without distinguishing their different perceptual importance, which brings redundancy in the learned codebook that not only limits the next stage’s autoregressive model’s ability to model important structure but also results in high training cost and slow generation speed. In this study, we borrow the idea of importance perception from classical image coding theory and propose a novel two-stage framework, which consists of Masked Quantization VAE (MQ-VAE) and Stackformer, to relieve the model from modeling redundancy. Specifically, MQ-VAE incorporates an adaptive mask module for masking redundant region features before quantization and an adaptive de-mask module for recovering the original grid image feature map to faithfully reconstruct the original images after quantization. Then, Stackformer learns to predict the combination of the next code and its position in the feature map. Comprehensive experiments on various image generation validate our effectiveness and efficiency. Code will be released at https: / / github. com / CrossmodalGroup / MaskedVectorQuantization. Copyright © 2023, The Authors. All rights reserved.

关键词： Redundancy

TRUST: A Toolkit for TEE-Assisted Secure Outsourced Computation over Integers

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhao, Bowen Li, Jiuhui Xu, Peiming Li, Xiaoguo Pei, Qingqi Shen, Yulong Guangzhou Institute of Technology Xidian University Guangzhou510555 China State Key Laboratory of Public Big Data Guizhou University Guiyang550025 China Electric Power Research Institute CSG Guangdong Guangzhou510663 China Guangdong Provincial Key Laboratory of Power System Network Security Guangdong Guangzhou510663 China The College of Computer Science Chongqing University Chongqing400044 China The State Key Laboratory of Integrated Service Networks Xidian University Xi’an710126 China

Secure outsourced computation (SOC) provides secure computing services by taking advantage of the computation power of cloud computing and the technology of privacy computing (e.g., homomorphic encryption). Expanding computational operations on encrypted data (e.g., enabling complex calculations directly over ciphertexts) and broadening the applicability of SOC across diverse use cases remain critical yet challenging research topics in the field. Nevertheless, previous SOC solutions frequently lack the computational efficiency and adaptability required to fully meet evolving demands. To this end, in this paper, we propose a toolkit for TEE-assisted (Trusted Execution Environment) SOC over integers, named TRUST. In terms of system architecture, TRUST falls in a single TEE-equipped cloud server only through seamlessly integrating the computation of REE (Rich Execution Environment) and TEE. In consideration of TEE being difficult to permanently store data and being vulnerable to attacks, we introduce a (2, 2)-threshold homomorphic cryptosystem to fit the hybrid computation between REE and TEE. Additionally, we carefully design a suite of SOC protocols supporting unary, binary and ternary operations. To achieve applications, we present SEAT, secure data trading based on TRUST. Security analysis demonstrates that TRUST enables SOC, avoids collusion attacks among multiple cloud servers, and mitigates potential secret leakage risks within TEE (e.g., from side-channel attacks). Experimental evaluations indicate that TRUST outperforms the state-of-the-art and requires no alignment of data as well as any network communications. Furthermore, SEAT is as effective as the Baseline without any data protection. Copyright © 2024, The Authors. All rights reserved.

关键词： Side channel attack

Adversarial Camouflage for Node Injection Attack on Graphs

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Tao, Shuchang Cao, Qi Shen, Huawei Wu, Yunfan Hou, Liang Sun, Fei Cheng, Xueqi Data Intelligence System Research Center Institute of Computing Technology Chinese Academy of Sciences Beijing China CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

Node injection attacks on Graph Neural Networks (GNNs) have received increasing attention recently, due to their ability to degrade GNN performance with high attack success rates. However, our study indicates that these attacks often fail in practical scenarios, since defense/detection methods can easily identify and remove the injected nodes. To address this, we devote to camouflage node injection attack, making injected nodes appear normal and imperceptible to defense/detection methods. Unfortunately, the non-Euclidean structure of graph data and the lack of intuitive prior present great challenges to the formalization, implementation, and evaluation of camouflage. In this paper, we first propose and define camouflage as distribution similarity between ego networks of injected nodes and normal nodes. Then for implementation, we propose an adversarial CAmouflage framework for Node injection Attack, namely CANA, to improve attack performance under defense/detection methods in practical scenarios. A novel camouflage metric is further designed under the guide of distribution similarity. Extensive experiments demonstrate that CANA can significantly improve the attack performance under defense/detection methods with higher camouflage or imperceptibility. This work urges us to raise awareness of the security vulnerabilities of GNNs in practical applications. Copyright © 2022, The Authors. All rights reserved.

关键词： Graph neural networks

Realizing Emotional Interactions to Learn User Experience and Guide Energy Optimization for Mobile Architectures 22

学校读者我要写书评

暂无评论

Realizing Emotional Interactions to Learn User Experience an...

Proceedings of the 55th Annual IEEE/ACM International Symposium on Microarchitecture

作者： Xueliang Li Zhuobin Shi Junyang Chen Yepang Liu National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China College of Computer Science and Software Engineering Shenzhen University China Research Institute of Trustworthy Autonomous Systems Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation and Department of Computer Science and Engineering Southern University of Science and Technology China

ISBN: (纸本)9781665462723

In the age of AI, mobile architectures such as smartphones are still "cold machines"; machines do not feel. If the architecture is able to feel users' feelings and runtime user experience (UX), it will accordingly adapt performance/energy to find the optimal system-operating state that consumes the least energy to satisfy users. In this paper, we will utilize users' facial expressions (FEs) to learn their runtime UX. We know that FEs are the natural and direct way for humans to convey their emotions and feelings. Our study reveals that FEs also reflect UX. Our research for the first time quantifies the link between FEs and UX. Leveraging this link, the architecture will be able to use the front camera to see FEs and feel users' UX. Based on UX, the architecture can appropriately provision computing resources. We propose Vi-energy system to realize the above idea. Our evaluation shows that Vi-energy reduces energy consumption by 52.9% at maximum and secures UX.

关键词：

Sequential end-to-end network for efficient person search

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Zhengjia Miao, Duoqian Department of Computer Science and Technology Tongji University Shanghai201804 China Key Laboratory of Embedded System and Service Computing Ministry of Education Shanghai201804 China

Person search aims at jointly solving Person Detection and Person Re-identification (re-ID). Existing works have designed end-to-end networks based on Faster R-CNN. However, due to the parallel structure of Faster R-CNN, the extracted features come from the low-quality proposals generated by the Region Proposal Network, rather than the detected high-quality bounding boxes. Person search is a fine-grained task and such inferior features will significantly reduce re-ID performance. To address this issue, we propose a Sequential End-to-end Network (SeqNet) to extract superior features. In SeqNet, detection and re-ID are considered as a progressive process and tackled with two sub-networks sequentially. In addition, we design a robust Context Bipartite Graph Matching (CBGM) algorithm to effectively employ context information as an important complementary cue for person matching. Extensive experiments on two widely used person search benchmarks, CUHK-SYSU and PRW, have shown that our method achieves state-of-the-art results. Also, our model runs at 11.5 fps on a single GPU and can be integrated into the existing end-to-end framework easily. Copyright © 2021, The Authors. All rights reserved.

关键词： Machine learning

Data protection and provenance in cloud of things environment: Research challenges

学校读者我要写书评

暂无评论

International Journal of Information and Computer Security 2020年第4期12卷 416-435页

作者： Wang, Chundong Yang, Lei Guo, Hao Wan, Fujin Key Laboratory of Computer Vision and System Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Ministry of Education Tianjin University of Technology Tianjin China Global Energy Internet Research Institute Beijing University of Technology Tianjin China Department of College of Computer and Control Engineering Nankai University Tianjin China

Internet of things are increasingly being deployed over the cloud (also referred to as cloud of things) to provide a broader range of services. However, there are serious challenges of CoT in the data protection and security provenance. This paper proposes a data privacy protection and provenance model (DPSPM) based on CoT. It can protect the privacy data of the users and trace the source of leaked data. In detail, security encryption and watermarking algorithms are proposed. Meanwhile, we use the improved k-anonymity data masking algorithm and pseudo-row watermarking algorithm in this scheme. Those algorithms can carry out security control over the whole process of data publishing, especially in data encryption, data masking and provenance verification. Finally, the experimental results show that our scheme has good efficiency. It is proved that the data masking time is proportional to the parameters k and L, the results also show good robustness to the common database watermarking attacks. © 2020 Inderscience Enterprises Ltd.

关键词： Data privacy

Merging Clinical Knowledge into Large Language Models for Medical Research and Applications: A Survey

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Li, Qiyuan Liu, Haijiang Guo, Caicai Chen, Deyu Wang, Meng Gao, Feng Gu, Jinguang College of Computer Science and Technology Wuhan University of Science and Technology Hubei Wuhan430065 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Hubei Wuhan430065 China The Key Laboratory of Rich-Media Knowledge Organization Service of Digital Publishing Content Institute of Scientific and Technical Information of China Beijing100038 China School of Computer Science and Technology Huazhong University of Science and Technology Hubei Wuhan430074 China School of Cyber Science and Engineering Wuhan University Hubei Wuhan430072 China

Clinical knowledge is the collection of information learned from studies on the causes, prognosis, diagnosis, and treatment of diseases. This type of knowledge can improve curing performances, and promote physical health. With the emergence of large language models (LLMs), medical artificial intelligence (medical AI), which aims to apply academic medical AI systems to real-world medical scenarios, has entered a new age of development, resulting in excellent works such as DoctorGPT and Pangu-Drug from academic and industrial researches. However, the field lacks a comprehensive compendium and comparison of building medical AI systems from academia and industry. Therefore, this survey focuses on the building paradigms of medical AI systems including the use of clinical databases, datasets, training pipelines, integrating medical knowledge graphs, system applications, and evaluation systems. We hope that this survey can help relevant practical researchers understand the current performance of academic models in various fields of healthcare, as well as the potential problems and future directions for implementing these scientific achievements. Copyright © 2025, The Authors. All rights reserved.

关键词： Clinical research

Gaussian Decay Function-based Improved Moment Matching for Ocular Artifacts Removal

学校读者我要写书评

暂无评论

Gaussian Decay Function-based Improved Moment Matching for O...

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Jiang, Hua Shi, Qiuxia Liu, Siying Zhang, Jiuying Zhao, Qinglin Hu, Bin Gansu Provincial Key Laboratory of Wearable Computing School of Information Science and Engineering China Lanzhou University Key Laboratory of Special Functional Materials and Structural Design Ministry of Education Lanzhou China CAS Center for Excellence in Brain Science and Intelligence Technology Shanghai Institutes for Biological Sciences Chinese Academy of Sciences China Joint Research Center for Cognitive Neurosensor Technology of Lanzhou University Institute of Semiconductors Chinese Academy of Sciences China Engineering Research Center of Open Source Software and Real-Time System Lanzhou University Ministry of Education China

ISBN: (纸本)9781665468190

Electroencephalogram (EEG) equipped with high time resolution that distracted by incoherent brain sources which including ocular artifacts (OAs) generating by blinks is of great significance. Therefore, these OAs got corrected before extracting information from EEG is indispensable. Improved Moment Matching (IMM) is a high-speed denoising algorithm suitable for removing OA in multi-channel EEG, which is an improvement of the moment matching method used to remove stripe noise in hyperspectral images. On foundation of this, this paper proposes an optimization algorithm for IMM based on Gaussian decay function (IMM_G). In the first place, the construction of the reference signal is optimized via utilizing a Gaussian decay function, thereby preventing the signal distortion caused by the filter. And then, a method for judging the to-be-processed interval is proposed to realizes the individualized processing of different channels. As a result, through quantitative comparison experiments with simulated data and real data from the UAIS laboratory, it was identified that IMM_G showed less time complexity, significantly enhanced arithmetic speed and denoising consequence while the detail retention ability of the original method for the non-blinking region of multi-channel EEG data is maintained. Hence, this method could be extensively used in High-density EEG (hdEEG) OAs removal scene. © 2022 IEEE.

关键词： Electroencephalography

Several classes of linear codes with few weights derived from Weil sums

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Hu, Zhao Qiu, Mingxiu Li, Nian Tang, Xiaohu Wu, Liwei Key Laboratory of Intelligent Sensing System and Security Hubei University Ministry of Education Hubei Provincial Engineering Research Center of Intelligent Connected Vehicle Network Security School of Cyber Science and Technology Hubei University Wuhan430062 China State Key Laboratory of Integrated Service Networks Xi’an710071 China Hubei Key Laboratory of Applied Mathematics Faculty of Mathematics and Statistics Hubei University Wuhan430062 China Southwest Jiaotong University Chengdu610031 China Wuhan Maritime Communication Research Institute China

Linear codes with few weights have applications in secret sharing, authentication codes, association schemes and strongly regular graphs. In this paper, several classes of t-weight linear codes over Fq are presented with the defining sets given by the intersection, difference and union of two certain sets, where t = 3, 4, 5, 6 and q is an odd prime power. By using Weil sums and Gauss sums, the parameters and weight distributions of these codes are determined completely. Moreover, three classes of optimal codes meeting the Griesmer bound are obtained, and computer experiments show that many (almost) optimal codes can be derived from our constructions. Copyright © 2025, The Authors. All rights reserved.

关键词： Weibull distribution