检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Cui, Shuang Li, Yi Li, Jiangmeng Tang, Xiongxin Su, Bing Xu, Fanjiang Xiong, Hui National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences China Beijing Key Laboratory of Big Data Management and Analysis Methods Gaoling School of Artificial Intelligence Renmin University of China China Department of Computer Science & Engineering the Hong Kong University of Science and Technology Hong Kong

Single image defocus deblurring (SIDD) aims to restore an all-in-focus image from a defocused one. Distribution shifts in defocused images generally lead to performance degradation of existing methods during out-of-distribution inferences. In this work, we gauge the intrinsic reason behind the performance degradation, which is identified as the heterogeneity of lens-specific point spread functions. Empirical evidence supports this finding, motivating us to employ a continual test-time adaptation (CTTA) paradigm for SIDD. However, traditional CTTA methods, which primarily rely on entropy minimization, cannot sufficiently explore task-dependent information for pixel-level regression tasks like SIDD. To address this issue, we propose a novel Siamese networks-based continual test-time adaptation framework, which adapts source models to continuously changing target domains only requiring unlabeled target data in an online manner. To further mitigate semantically erroneous textures introduced by source SIDD models under severe degradation, we revisit the learning paradigm through a structural causal model and propose Causal Siamese networks (CauSiam). Our method leverages large-scale pre-trained vision-language models to derive discriminative universal semantic priors and integrates these priors into Siamese networks, ensuring causal identifiability between blurry inputs and restored images. Extensive experiments demonstrate that CauSiam effectively improves the generalization performance of existing SIDD methods in continuously changing domains. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

ON CHARACTERIZING THE INTER-DEPARTURE PROCESS OF A SERVER

引用

PERFORMANCE EVALUATION 1982年第4期2卷 251-255页

作者： TRIPATHI, SK AGRAWALA, AK System Design and Analysis Group Department of Computer Science University of Maryland College Park MD 20742 U.S.A.

A technique for computing the inter-departure time distribution for single server queues is presented. The method is applicable for both the steady state and the transient characterizations. Examples of its use in several situations are presented. An explicit expression for the interdeparture time distribution of a steady state G/M/1 queue is obtained.

关键词： Departure Process Single Server Queue Virtual Waiting Time Sojourn Time Inter-Departure Distribution Characterization of Transient Inter-Departure Distribution

来源：评论

学校读者我要写书评

暂无评论

A Paraphrasing system for transforming regular expressions into honorifics

A Paraphrasing system for transforming regular expressions i...

引用

23rd International Conference on Computational Linguistics, Coling 2010

作者： Han, Dongli Kamochi, Shuntaro Song, Xin Akegawa, Naoki Hori, Tomomasa Department of Computer Science and System Analysis College of Humanities and Sciences Nihon University Japan

Honorifics in Japanese plays an incredibly important role in all walks of social life. The demand to transform regular expressions in Japanese into honorifics automatically has increased rapidly especially in business situations. This paper reviews existing studies and proposes a system to fill this demand with more practicable functions. The experiment shows the effectiveness of our strategy.

关键词： Pattern matching

来源：评论

学校读者我要写书评

暂无评论

A knowledge-based system for measuring the quality of the distribution of salami slices on pizza crusts

A knowledge-based system for measuring the quality of the di...

引用

2011 1st IEEE Jordan Conference on Applied Electrical Engineering and Computing Technologies, AEECT 2011

作者： Mashi, Mohammad Ament, Christoph Department of System Analysis Faculty of Computer Science and Automation Ilmenau University of Technology Ilmenau Germany

ISBN: (纸本)9781457710841

All manufacturing companies have been trying to integrate intelligence-oriented strategies into their workflow. This saves time and money and leads to a better quality of goods. This intelligent strategy is in the subject area of Artificial Intelligence (AI). In our contribution a Knowledge-Based system (KBS) for image processing and pattern recognition, machine learning (ML) is used to measure the quality of the distribution of salami slices on pizza crusts. Measuring quality provides the opportunity to correct the position of the salami slices by automated systems (e.g. a robot arm) in the future rather than by employees. © 2011 IEEE.

关键词： Fuzzy logic

来源：评论

学校读者我要写书评

暂无评论

Automatic utterance generation by keeping track of the conversation's focus within the utterance window

Automatic utterance generation by keeping track of the conve...

引用

8th International Conference on Natural Language Processing, JapTAL 2012

作者： Nishio, Yusuke Han, Dongli Department of Computer Science and System Analysis College of Humanities and Sciences Nihon University Tokyo Japan

ISBN: (纸本)9783642339820

The insufficiency in methods for generating utterances still remains as a critical issue unsolved in the community of non-task-oriented conversation. Previous studies provide various strategies to enrich the methods for generating utterances, thus making the conversation systems or agents appear more interesting. However, none of them could escape from the fact that they all generate utterances depending mainly on some particular kinds of templates or augmented templates. We propose here in this paper a thorough modification to a preceding work to address this problem. Specifically, we first introduce a concept Utterance Window to strengthen the association between continuous utterances, and then employ a Two-starting-word Markov connection to cope with the ease of losing focus of the current utterance. In addition, we try to keep tract of user's interests and reflecting them in the process of topic-word extraction and utterance generation as well. The experimental results show the effectiveness of our method. © 2012 Springer-Verlag Berlin Heidelberg.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Parallel quasi-Monte Carlo integration using (t,s)-sequences 4th

引用

4th International ACPC Conference on Parallel Computation, ACPC 1999

作者： Schmid, Wolfgang Ch. Uhl, Andreas Department of Mathematics University of Salzburg Austria RIST++ and Department of Computer Science and System Analysis University of Salzburg Austria

ISBN: (纸本)3540656413

Currently, the most effective constructions of low-discrepancy point sets and sequences are based on the theory of (t, m, s)-nets and (t, s)-sequences. In this work we discuss parallelization techniques for quasi-Monte Carlo integration using (t, s)-sequences. We show that leapfrog parallelization may be very dangerous whereas block-based parallelization turns out to be robust. © Springer-Verlag Berlin Heidelberg 1999.

关键词： Monte Carlo methods

来源：评论

学校读者我要写书评

暂无评论

DESIGNING FAULT TOLERANT APPLICATIONS IN MARUTI 3

DESIGNING FAULT TOLERANT APPLICATIONS IN MARUTI

引用

3RD INTERNATIONAL SYMPOSIUM ON SOFTWARE RELIABILITY ENGINEERING

作者： LIANG, DR AGRAWALA, A MOSSE, D SHI, YH System Design and Analysis Group Computer Science Department University of Maryland at College Park College Park 20742 MD United States

ISBN: (纸本)0818629754

In this paper, we introduce a model for developing applications with fault-tolerance requirements and real-time constraints. Applications in this model are specified using computation graphs, in which vertices represent tasks and arcs represent precedence constraints. Tasks are replicated to provide required fault tolerance and ensure that a real-time application will meet its deadlines despite failures. We develop an analytical model to calculate the probability of successful execution of applications with task replications. We propose an efficient algorithm for the analysis of applications that are composed of subgraphs, each of which has a single source and a single sink. The results of the analysis can be used with information from allocation/scheduling to develop applications with desired timing and fault-tolerance requirements. © 1992 IEEE.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Parallel computation of optimal parameters for pseudo random number generation 3rd

Parallel computation of optimal parameters for pseudo random...

引用

3rd International Austrian Center for Parallel Computation Conference with Special Emphasis on Parallel Databases and Parallel I/O, 1996

作者： Brunner, D. Uhl, A. Department of Mathematics Salzburg University Austria Research Institute for Softwaretechnology and Department of Computer Science and System Analysis Salzburg University Austria

ISBN: (纸本)3540616950

Two systematic search methods are employed to find multipliers for linear congruential pseudo-random number generation which are optimal with respect to the discrepancy of pairs of successive pseudo-random numbers. These two methods are compared in terms of their suitability for parallel computation. Experimental results of a MIMD workstationcluster-implementation and an evaluation of the calculated parameters using the spectral-test are included. © 9 Springer-Verlag Berlin Heidelberg 1996.

关键词： Parameter estimation

来源：评论

学校读者我要写书评

暂无评论

On updating buffer allocation

On updating buffer allocation

引用

computer Network Performance Symposium 1982

作者： Thareja, Ashok K. Tripathi, Satish K. Upton, Richard A. System Design and Analysis Group Department of Computer Science University of Maryland College ParkMD20742 United States

ISBN: (纸本)0897910699

Most of the analysis of buffer sharing schemes has been aimed at obtaining the optimal operational parameters under stationary load situations. It is well known that in most operating environments the traffic load changes. In this paper, we address the problem of updating buffer allocation as the traffic load at a network node changes. We investigate the behavior of a complete partitioning buffer sharing scheme to gain insight into the dependency of the throughput upon system parameters. The summary of the analysis is presented in the form of a heuristic. The heuristic is shown to perform reasonably well under two different types of stress tests. © 1982 ACM.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Parallel algorithms for using non-stationary MRA in image compression 2nd

引用

2nd International Euro-Par Conference on Parallel Processing, Euro-Par 1996

作者： Uhl, Andreas Research Institute for Softwaretechnology Department of Computer Science and System Analysis Salzburg University Hellbrunnerstr.34 Salzburg5020 Austria

ISBN: (纸本)3540616276

New methods for compact image coding based on generalized wavelet decompositions have been introduced recently. Unlike in the classical wavelet decomposition scheme it is possible to use different scaling and wavelet functions at every scale by using non-stationary multiresolution analyses. In this work we introduce parallel algorithms (suitable for MIMD architectures) that excel the execution speed for this type of lossy compression algorithms by far. © 1996, Springer Verlag. All rights reserved.

关键词： Wavelet decomposition

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：