检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Zhao, Minyi Xu, Yi Li, Bingjia Wang, Jie Guan, Jihong Zhou, Shuigeng Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University China ByteDance Inc China Department of Computer Science and Technology Tongji University China

Scene text image super-resolution (STISR) is an important pre-processing technique for text recognition from low-resolution scene images. Nowadays, various methods have been proposed to extract text-specific information from high-resolution (HR) images to supervise STISR model training. However, due to uncontrollable factors (e.g. shooting equipment, focus, and environment) in manually photographing HR images, the quality of HR images cannot be guaranteed, which unavoidably impacts STISR performance. Observing the quality issue of HR images, in this paper we propose a novel idea to boost STISR by first enhancing the quality of HR images and then using the enhanced HR images as supervision to do STISR. Concretely, we develop a new STISR framework, called High-Resolution ENhancement (HiREN) that consists of two branches and a quality estimation module. The first branch is developed to recover the low-resolution (LR) images, and the other is an HR quality enhancement branch aiming at generating high-quality (HQ) text images based on the HR images to provide more accurate supervision to the LR images. As the degradation from HQ to HR may be diverse, and there is no pixel-level supervision for HQ image generation, we design a kernel-guided enhancement network to handle various degradation, and exploit the feedback from a recognizer and text-level annotations as weak supervision signal to train the HR enhancement branch. Then, a quality estimation module is employed to evaluate the qualities of HQ images, which are used to suppress the erroneous supervision information by weighting the loss of each image. Extensive experiments on TextZoom show that HiREN can work well with most existing STISR methods and significantly boost their performances. © 2024, The Authors. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Edge-Assisted Adaptive Configuration for Serverless-Based Video Analytics

Edge-Assisted Adaptive Configuration for Serverless-Based Vi...

引用

International Conference on Distributed Computing Systems

作者： Ziyi Wang Songyu Zhang Jing Cheng Zhixiong Wu Zhen Cao Yong Cui Department of Computer Science and Technology Tsinghua University China Computer Network and Protocol Lab Huawei Technologies China

The growth of video volumes and increased DNN capabilities have led to a growing desire for video analytics, which demands intensive computation resources. Traditional resource provisioning strategies, such as configuring a cluster per peak utilization, lead to low resource efficiency. Serverless computing is a promising way to avoid wasteful resource provisioning since video analytics regularly encounters bursty input workloads and finegrained video content dynamics. For serverless-based video analytics, the application configuration (frame rate, detection model, and computation resources) will impact several metrics, such as computation cost and analytics accuracy. In this paper, we investigate the joint configuration adjustment problem for video knobs and computation resources provided by the serverless platform. We propose an algorithm that can efficiently adapt configurations for video streams to address two key challenges in serverless-based video analytics systems, including the complex relationships between the configurations and the key performance metrics, and the dynamically best configuration. Our algorithm is developed based on Markov approximation to minimize the computation cost within an accuracy constraint. We have developed a prototype over AWS Lambda and conducted extensive experiments with real-world video streams. The results show that our algorithm can greatly reduce the computation cost under the constraint of target accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing LLM Reliability via Explicit Knowledge Boundary Modeling

arXiv

引用

arXiv 2025年

作者： Zheng, Hang Xu, Hongshen Liu, Yuncong Chen, Lu Fung, Pascale Yu, Kai X-LANCE Lab Department of Computer Science and Engineering Shanghai Jiao Tong University China Hong Kong University of Science and Technology Hong Kong

Large language models (LLMs) frequently hallucinate due to misaligned self-awareness, generating erroneous outputs when addressing queries beyond their knowledge boundaries. While existing approaches mitigate hallucinations via uncertainty estimation or query rejection, they suffer from computational inefficiency or sacrificed helpfulness. To address these issues, we propose the Explicit Knowledge Boundary Modeling (EKBM) framework, integrating fast and slow reasoning systems to harmonize reliability and usability. The framework first employs a fast-thinking model to generate confidence-labeled responses, enabling immediate use of high-confidence outputs. For uncertain predictions, a slow refinement model conducts targeted reasoning to improve accuracy. To align model behavior with our proposed object, we propose a hybrid training pipeline, enhancing self-awareness without degrading task performance. Evaluations on dialogue state tracking tasks demonstrate that EKBM achieves superior model reliability over uncertainty-based baselines. Further analysis reveals that refinement substantially boosts accuracy while maintaining low computational overhead. Our work establishes a scalable paradigm for advancing LLM reliability and balancing accuracy and practical utility in error-sensitive applications. Copyright © 2025, The Authors. All rights reserved.

关键词： Reliability analysis

来源：评论

学校读者我要写书评

暂无评论

Cognitive Skill Enhancement System Using Neuro-Feedback for ADHD Patients

引用

computers, Materials & Continua 2021年第8期68卷 2363-2376页

作者： Muhammad Usman Ghani Khan Zubaira Naz Javeria Khan Tanzila Saba Ibrahim Abunadi Amjad Rehman Usman Tariq Al-khawarizmi Institute of Computer Science University of Engineering and TechnologyLahore54890Pakistan Department of Computer Science University of Engineering and TechnologyLahore54890Pakistan Artificial Intelligence and Data Analytics Lab CCIS Prince Sultan UniversityRiyadhSaudi Arabia College of Computer Engineering and Science Prince Sattam bin Abdulaziz UniversityAlkharjSaudi Arabia

The National Health Interview Survey(NHIS)shows that there are 13.2%of children at the age of 11 to 17 who are suffering from Attention Deficit Hyperactivity Disorder(ADHD),*** treatment methods for ADHD are either psycho-stimulant medications or cognitive *** traditional methods,namely therapy,need a large number of visits to hospitals and include *** could be used for the effective treatment of *** could be a helpful tool in improving children and ADHD patients’cognitive skills by using Brain–computer Interfaces(BCI).BCI enables the user to interact with the computer through brain activity using Electroencephalography(EEG),which can be used to control different computer applications by processing acquired brain *** paper proposes a system based on neurofeedback that can improve cognitive skills such as attention level,mediation level,and spatial *** proposed system consists of a puzzle game where its complexity increases with each *** signals were acquired using the Neurosky headset;then sent the signals to the designed gaming *** neurofeedback system was tested on 10 different subjects,and their performance was calculated using different evaluation *** results show that this game improves player overall performance from 74%to 98%by playing each game level.

关键词： Neurogaming brain-computer interface attention deficit hyperactivity disorder electroencephalography spatial memory cognitive skills attention level

来源：评论

学校读者我要写书评

暂无评论

DEEP GENERATIVE MODELING ON LIMITED DATA WITH REGULARIZATION BY NONTRANSFERABLE PRE-TRAINED MODELS 11

DEEP GENERATIVE MODELING ON LIMITED DATA WITH REGULARIZATION...

引用

11th International Conference on Learning Representations, ICLR 2023

作者： Zhong, Yong Liu, Hongtao Liu, Xiaodong Bao, Fan Shen, Weiran Li, Chongxuan Gaoling School of AI Renmin University of China Beijing China Beijing Key Lab of Big Data Management and Analysis Methods Beijing China Department of Computer Science Technology Tsinghua University Beijing China

Deep generative models (DGMs) are data-eager because learning a complex model on limited data suffers from a large variance and easily overfits. Inspired by the classical perspective of the bias-variance tradeoff, we propose regularized deep generative model (Reg-DGM), which leverages a nontransferable pre-trained model to reduce the variance of generative modeling with limited data. Formally, Reg-DGM optimizes a weighted sum of a certain divergence and the expectation of an energy function, where the divergence is between the data and the model distributions, and the energy function is defined by the pre-trained model w.r.t. the model distribution. We analyze a simple yet representative Gaussian-fitting case to demonstrate how the weighting hyperparameter trades off the bias and the variance. Theoretically, we characterize the existence and the uniqueness of the global minimum of Reg-DGM in a non-parametric setting and prove its convergence with neural networks trained by gradient-based methods. Empirically, with various pre-trained feature extractors and a data-dependent energy function, Reg-DGM consistently improves the generation performance of strong DGMs with limited data and achieves competitive results to the state-of-the-art methods. Our implementation is available at https://***/ML-GSAI/Reg-ADA-APA. © 2023 11th International Conference on Learning Representations, ICLR 2023. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DeltaBin: An Efficient Binary Data Format for Low Power IoT Devices

DeltaBin: An Efficient Binary Data Format for Low Power IoT ...

引用

International Conference on computer, Information and Telecommunication Systems (CITS)

作者： Neda Maleki Arslan Musaddiq David Mozart Tobias Olsson Mustafa Omareen Fredrik Ahlgren Department of Computer Science and Media Technology Applied IoT Lab Linnaeus University Sweden

The Internet of Things (IoT) notion is quickly influencing t he architectures of data-driven systems d ue to the ever-increasing rapid technological progress in all sectors. The IoT involves the collection and exchange of data from a large number of interconnected devices or sensors. The collected data is structured and transmitted in a variety of different data formats such as JSON, CBOR, BSON, or simply a binary format. The data format used by an IoT device can have a significant i mpact on t he efficiency of its data transmission. In general, using a more compact and efficient data format can help to reduce t he amount of data that needs to be transmitted, which can improve the overall speed and performance of the device. For example, using a binary data format rather than a text-based format can often result in smaller data sizes and faster transmission times. Similarly, using a binary format in a more compressed form can further help to reduce the size of the data being transmitted, which can further improve the efficiency of the transmission. In this paper, we propose Delta Binary (i.e., DeltaBin) to reduce the binary data format by transmitting only changed data. We assess DeltaBin using a real IoT deployment scenario.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Integrating Object Detection and Wide Area Network Infrastructure for Sustainable Ferry Operation

Integrating Object Detection and Wide Area Network Infrastru...

引用

IEEE International Workshop on Imaging Systems and Techniques (IST)

作者： Arslan Musaddiq David Mozart Neda Maleki Tobias Olsson Fredrik Ahlgren Department of Computer Science and Media Technology Applied IoT Lab Linnaeus University Sweden

Low-Power Wide-Area Network (LPWAN) technologies offer new opportunities for data collection, transmission, and decision-making optimization. Similarly, a wide range of use cases of computer vision and object detection algorithms can be found across different industries. This paper presents a case study focusing on the utilization of LPWAN infrastructure, specifically the Helium network, coupled with computer vision and object detection algorithms, to optimize passenger ferry operation. The passenger ferry called M/S Dessi operates between Kalmar and Färjestaden in Sweden during the summer season. By implementing an Edge-computing solution, real-time data collection and communication are achieved, enabling accurate measurement of passenger flow. This approach is superior to traditional methods of collecting passenger data, such as manual counting or CCTV surveillance. Real-time passenger data is invaluable for traffic planning, crowd prediction, revenue enhancement, and speed and fuel optimization. The utilization of the Helium network ensures reliable and long-distance data transmission, extending the system’s applicability to multiple ferries and distant locations. The proposed approach can be utilized to integrate passenger ferries that operate in close proximity to urban areas into society’s digital transformation efforts. This study highlights the potential of LPWAN, computer vision, and object detection in enhancing passenger ferry operations, contributing to enhanced efficiency and sustainability.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Nothing Stands Alone: Relational Fake News Detection with Hypergraph Neural Networks

Nothing Stands Alone: Relational Fake News Detection with Hy...

引用

2022 IEEE International Conference on Big Data, Big Data 2022

作者： Jeong, Ujun Ding, Kaize Cheng, Lu Guo, Ruocheng Shu, Kai Liu, Huan Arizona State University School of Computing and Augmented Intelligence United States University of Illinois Chicago Department of Computer Science United States Bytedance Ai Lab London United Kingdom Illinois Institute of Technology Department of Computer Science United States

ISBN: (纸本)9781665480451

Nowadays, fake news easily propagates through online social networks and becomes a grand threat to individuals and society. Assessing the authenticity of news is challenging due to its elaborately fabricated contents, making it difficult to obtain large-scale annotations for fake news data. Due to such data scarcity issues, detecting fake news tends to fail and overfit in the supervised setting. Recently, graph neural networks (GNNs) have been adopted to leverage the richer relational information among both labeled and unlabeled instances. Despite their promising results, they are inherently focused on pairwise relations between news, which can limit the expressive power for capturing fake news that spreads in a group-level. For example, detecting fake news can be more effective when we better understand relations between news pieces shared among susceptible users. To address those issues, we propose to leverage a hypergraph to represent group-wise interaction among news, while focusing on important news relations with its dual-level attention mechanism. Experiments based on two benchmark datasets show that our approach yields remarkable performance and maintains the high performance even with a small subset of labeled news data. © 2022 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

An Exploratory Study on Energy Consumption of Dataframe Processing Libraries

An Exploratory Study on Energy Consumption of Dataframe Proc...

引用

International Mining Software Repositories, MSR

作者： Shriram Shanbhag Sridhar Chimalakonda Department of Computer Science & Engineering RISHA Lab Indian Institute of Technology Tirupati India

The energy consumption of machine learning applications and their impact on the environment has recently gained attention as a research area, focusing on the model creation and training/inference phases. The data-oriented stages of the machine learning pipeline, which involve pre-processing, cleaning, and exploratory analysis, are critical components. However, energy consumption during these stages has received limited attention. Dataframe processing libraries play a significant role in these stages, and optimizing their energy consumption is important for reducing environmental impact and operational costs. Therefore, as a first step towards studying their energy efficiency, we investigate and compare the energy consumption of three popular dataframe processing libraries, namely Pandas, Vaex, and Dask. We perform experiments across 21 dataframe processing operations within four categories, utilizing three distinct datasets. Our results indicate that no single library is the most energy-efficient for all tasks, and the choice of a library can have a significant impact on energy consumption based on the types and frequencies of operations performed. The findings of this study suggest the potential for optimization of the energy consumption of data-oriented stages in the machine learning pipeline and warrant further research in this area.

关键词：

来源：评论

学校读者我要写书评

暂无评论

On the necessity of collaboration for online model selection with decentralized data 24

On the necessity of collaboration for online model selection...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Junfan Li Zheshun Wu Zenglin Xu Irwin King School of Computer Science and Technology Harbin Institute of Technology Shenzhen Pengcheng Lab and Artificial Intelligence Innovation and Incubation (AI 3 ) Institute Fudan University and Shanghai Academy of AI for Science Department of Computer Science and Engineering The Chinese University of Hong Kong

ISBN: (纸本)9798331314385

We consider online model selection with decentralized data over M clients, and study the necessity of collaboration among clients. Previous work proposed various federated algorithms without demonstrating their necessity, while we answer the question from a novel perspective of computational constraints. We prove lower bounds on the regret, and propose a federated algorithm and analyze the upper bound. Our results show (i) collaboration is unnecessary in the absence of computational constraints on clients; (ii) collaboration is necessary if the computational cost on each client is limited to o(K), where K is the number of candidate hypothesis spaces. We clarify the unnecessary nature of collaboration in previous federated algorithms for distributed online multi-kernel learning, and improve the regret bounds at a smaller computational and communication cost. Our algorithm relies on three new techniques including an improved Bernstein's inequality for martingale, a federated online mirror descent framework, and decoupling model selection and prediction, which might be of independent interest.

关键词：

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：