检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Wang, Xiaoqin Ma, Xusen Hou, Xianxu Ding, Meidan Li, Yudong Chen, Junliang Chen, Wenting Peng, Xiaoyang Shen, Linlin Computer Vision Institute College 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 AIAC Xi’an Jiaotong-Liverpool University China Tsinghua University China The Hong Kong Polytechnic University Hong Kong City University of Hong Kong Hong Kong Sun Yat-sen University China

Multimodal large language models (MLLMs) have demonstrated remarkable capabilities in various tasks. However, effectively evaluating these MLLMs on face perception remains largely unexplored. To address this gap, we introduce FaceBench, a dataset featuring hierarchical multi-view and multi-level attributes specifically designed to assess the comprehensive face perception abilities of MLLMs. Initially, we construct a hierarchical facial attribute structure, which encompasses five views with up to three levels of attributes, totaling over 210 attributes and 700 attribute values. Based on the structure, the proposed FaceBench consists of 49,919 visual question-answering (VQA) pairs for evaluation and 23,841 pairs for fine-tuning. Moreover, we further develop a robust face perception MLLM baseline, Face-LLaVA, by training with our proposed face VQA data. Extensive experiments on various mainstream MLLMs and Face-LLaVA are conducted to test their face perception ability, with results also compared against human performance. The results reveal that, the existing MLLMs are far from satisfactory in understanding the fine-grained facial attributes, while our Face-LLaVA significantly outperforms existing open-source models with a small amount of training data and is comparable to commercial ones like GPT-4o and Gemini. The dataset will be released at https://***/CVI-SZU/FaceBench. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Active transfer learning of matching query results across multiple sources

引用

Frontiers of computer Science 2015年第4期9卷 595-607页

作者： Jie XIN Zhiming CUI Pengpeng ZHAO Tianxu HE The Institute of Intelligent Information Processing and Application Soochow University Suzhou 215006 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou 215006 China

Entity resolution （ER） is the problem of identi- fying and grouping different manifestations of the same real world object. Algorithmic approaches have been developed where most tasks offer superior performance under super- vised learning. However, the prohibitive cost of labeling training data is still a huge obstacle for detecting duplicate query records from online sources. Furthermore, the unique combinations of noisy data with missing elements make ER tasks more challenging. To address this, transfer learning has been adopted to adaptively share learned common structures of similarity scoring problems between multiple sources. Al- though such techniques reduce the labeling cost so that it is linear with respect to the number of sources, its random sam- piing strategy is not successful enough to handle the ordinary sample imbalance problem. In this paper, we present a novel multi-source active transfer learning framework to jointly select fewer data instances from all sources to train classi- fiers with constant precision/recall. The intuition behind our approach is to actively label the most informative samples while adaptively transferring collective knowledge between sources. In this way, the classifiers that are learned can be both label-economical and flexible even for imbalanced or quality diverse sources. We compare our method with the state-of-the-art approaches on real-word datasets. Our exper- imental results demonstrate that our active transfer learning algorithm can achieve impressive performance with far fewerlabeled samples for record matching with numerous and var- ied sources.

关键词： entity resolution active learning transfer learn-ing convex optimization

来源：评论

学校读者我要写书评

暂无评论

Ensemble feature selection integrating elitist roles and quantum game model

引用

Journal of Systems Engineering and Electronics 2015年第3期26卷 584-594页

作者： Weiping Ding Jiandong Wang Zhijin Guan Quan Shi College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics School of Computer Science and Technology Nantong University Provincial Key Laboratory for Computer Information Processing Technology Soochow University

To accelerate the selection process of feature subsets in the rough set theory （RST）, an ensemble elitist roles based quantum game （EERQG） algorithm is proposed for feature selec- tion. Firstly, the multilevel elitist roles based dynamics equilibrium strategy is established, and both immigration and emigration of elitists are able to be self-adaptive to balance between exploration and exploitation for feature selection. Secondly, the utility matrix of trust margins is introduced to the model of multilevel elitist roles to enhance various elitist roles＇ performance of searching the optimal feature subsets, and the win-win utility solutions for feature selec- tion can be attained. Meanwhile, a novel ensemble quantum game strategy is designed as an intriguing exhibiting structure to perfect the dynamics equilibrium of multilevel elitist roles. Finally, the en- semble manner of multilevel elitist roles is employed to achieve the global minimal feature subset, which will greatly improve the fea- sibility and effectiveness. Experiment results show the proposed EERQG algorithm has superiority compared to the existing feature selection algorithms.

关键词： ensemble quantum game utility matrix of trust mar-gin dynamics equilibrium strategy multilevel elitist role feature selection and classification.

来源：评论

学校读者我要写书评

暂无评论

Privacy preservation in role-based access control model

引用

Journal of Networks 2011年第8期6卷 1106-1113页

作者： Chen, Zuo Yang, Qiuwei Wan, Xin Tu, Yuanyan Yu, Fei Xu, Chen School of Computer and Communication Hunan University Changsha China Jiangsu Provincial Key Laboratory for Computer Information Processing Technology Soochow University Soochow China

Privacy preservation is a crucial problem in resource sharing and collaborating among multi-domains. Based on this problem, we propose a role-based access control model for privacy preservation. This scheme avoided the privacy leakage of resources while implementing access control, and it has the advantage of lower communication overhead. We demonstrate this scheme meets the IND-CCA2 semantic security by using random oracle. The simulation result shows this scheme has better execution efficiency and application effects. © 2011 ACADEMY PUBLISHER.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Developing software using a novel event-condition-state pattern framework: Taking mobile input method system As A Case Study

Journal of Software

引用

Journal of Software 2011年第8期6卷 1506-1512页

作者： Zhu, Fei Diao, Hongjun Huang, Wei School of Computer Science and Technology Soochow University Suzhou Jiangsu 215006 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou Jiangsu 215006 China

Design patterns add more reliability, flexibility and reusability to a software system. Taking advantage of design patterns is usually beneficial to software design and makes software development relatively easier. The state pattern proposed by GoF is not a determined design and can hardly be transformed into implementation code directly. It is disable to cope with sophisticated response event either. Hereby we propose a novel event-conditionstate pattern framework, an event-based finite state machine. The framework improves GoF state pattern by resegmenting Context into two classes, one for management and the other for operation interface. Essential components for implementation, such as management, state transferring, and event triggering mechanism, are also taken into consideration in the framework. We use pattern framework to develop a thirty party input method system which has a Windows Mobile system version and a Symbian system version separately. We rewrote Context part related with system interfaces in the two different systems, and encapsulated the part related with internal logic of input method by concrete subclasses, thus avoiding redundant developing work in both systems and providing convenience for later synchronous updating and maintenance of two input methods. With the pattern framework, much effort during software development is saved. The development logic seems simpler and clearer. As a result, the whole development process is simplified. © 2011 ACADEMY PUBLISHER.

关键词： Software design

来源：评论

学校读者我要写书评

暂无评论

Probabilistic fault diagnosis of clustered faults for multiprocessor systems

引用

Journal of computer Science & technology 2023年第4期38卷 821-833页

作者：孙雪丽樊建席程宝雷王岩张力 School of Computer Science and Technology Soochow UniversitySuzhou 215006China Provincial Key Laboratory for Computer Information Processing Technology Soochow UniversitySuzhou 215006China

With the development of high-performance computing and the expansion of large-scale multiprocessor sys-tems,it is significant to study the reliability of *** fault diagnosis is of practical value to the reliabilityanalysis of multiprocessor *** this paper,we design a linear time diagnosis algorithm with the multiprocessor sys-tem whose threshold is set to 3,where the probability that any node is correctly diagnosed in the discrete state can be ***,we give the probabilities that all nodes of a d-regular and d-connected graph can be correctly diag-nosed in the continuous state under the Weibull fault distribution and the Chi-square fault *** prove thatthey approach to 1,which implies that our diagnosis algorithm can correctly diagnose almost all nodes of the graph.

关键词： regular graph clustered fault probabilistic diagnosis algorithm Preparata Metze Chien model(PMCmodel)

来源：评论

学校读者我要写书评

暂无评论

A Topic-Triggered Translation Model for Statistical Machine Translation

引用

Chinese Journal of Electronics 2017年第1期26卷 65-72页

作者： SU Jinsong WANG Zhihao WU Qingqiang YAO Junfeng LONG Fei ZHANG Haiying Software School Xiamen University Provincial Key Laboratory for Computer Information Processing Technology Soochow University Automation Department Xiamen University

Translation model containing translation rules with probabilities plays a crucial role in statistical machine translation. Conventional method estimates translation probabilities with only the consideration of cooccurrence frequencies of bilingual translation units, while ignoring document-level context information. In this paper, we extend the conventional translation model to a topic-triggered one. Specifically, we estimate topic-specific translation probabilities of translation rules by leveraging topical context information, and online score selected translation rules according to topic posterior distributions of translated sentences. As compared with the conventional model, our model allows for more fine-grained distinction among different translations. Experiment results on large data set demonstrate the effectiveness of our model.

关键词： Statistical machine translation Topictriggered translation model Topical context information

来源：评论

学校读者我要写书评

暂无评论

Enriched format text categorization using a component similarity approach

Journal of Software

引用

Journal of Software 2011年第9期6卷 1713-1720页

作者： Zhu, Fei Yang, Jiong Zhou, Yong School of Computer Science and Technology Soochow University Suzhou Jiangsu 215006 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou China

Text categorization has been widely studied for years. However, conventional plain text categorization approaches which work good in plain text behave poor when they are simply applied to enriched format texts. An categorization approach that is applicable to enriched format text is proposed. During feature selection, we get feature structure distribution weight by using extended structure model so that structure affections to categorization are fully considered. Text formats are also taken into account in feature weighting. The combined feature weighting approach strengthens important parts and weakens less important ones. The text categorization is fulfilled by document component similarity, which first decomposes document, gathers features by components and other user-defined rules, completes document component tree, and then achieves categorization by it. We implement a CSBC based Naïve Bayes classifier in which the final result is the combination of all classifiers of component tree. Finally we parse *** document, draw components that are most related to classification from *** documents, and then use the classifier to categorize *** documents. The experiment results show that the classifier can automatically classify *** documents and work quite well. © 2011 ACADEMY PUBLISHER.

关键词： Classifiers

来源：评论

学校读者我要写书评

暂无评论

Implementation of data-exchanging system based on message oriented middleware in agricultural website

引用

WSEAS Transactions on computers 2008年第6期7卷 620-629页

作者： Zhang, Xiaoshuan Wu, Qinghua Tian, Dong Zhao, Ming Xiaoshuan, Zhang College of Information and Electrical Engineering China Agricultural University Beijing 100083 China Jiangsu Provincial Key Laboratory Computer Information Processing Technology Suzhou China

With the startup of the Golden Agriculture Project, the step of being-information of agriculture is becoming rapid. And the transformation and share of data is indispensable to the being-information of agriculture. How to implement data combination, data transformation and data receiving applications are the important means to complete the information share safely and enhance the efficiency. The paper starts with searching of methods to implement data interchange, and introduce some of the methods, points of the techniques, etc. Basing on this, the paper also introduces the detail requirement analyses, system design and detail implementation of the system. According to the requirement and trait of the project, a data interchange system is researched and completed. And a data interchange model based on message-oriented middleware (MOM) is presented in this paper, which builds a middleware between the province and the ministry taking part in data interchange. The system has traits as follows: 1. keeping the data safe and credible while it is transformed. 2. having excellent transplantable and applied capability. 3. doesn't need intervention of workman in the process of data interchange. 4. applying the data interchange between databases of different structure. 5. being simple to be developed and applied. MOM TongLink/Q offers interfaces for application development, and it completes the data transformation through the internet. The integration adapters developed do the data management, which are developed based on the Frame for Applications Integration TongIntegrator. This method offers a new approach to resolve the question of data interchange. Now the system has been successfully applied in the data interchange project of Ministry of Agriculture.

关键词： Electronic data interchange

来源：评论

学校读者我要写书评

暂无评论

Automated trust negotiation with time behavior

引用

Journal of Networks 2011年第8期6卷 1091-1098页

作者： Chen, Zuo Yang, Qiuwei Wan, Xin Tu, Yuanyan Yu, Fei Xu, Chen School of Computer and Communication Hunan University Chang sha China Jiangsu Provincial Key Laboratory for Computer Information Processing Technology Soochow University Soochow China

In automated trust negotiation(ATN), strangers build trust relationship by disclosing attributes credentials alternately. In the recent study of ATN, there was no rigorous formal definition of ATN abstract model, and have nothing to do with the time behavior. It exist denial of service attack which can lead system to paralysis. We present the basic components of ATN abstract model. It defines ATN abstract model by bringing state-transition system. Based on this formalized basis, the time behavior will be explored. We analyze and discuss some time semantics of ATN, and describe the extended ATN by constructing a state-transition system with time behavior. It can describe the ATN abstract model more comprehensive and secure. The satiability of security policies in ATN is discussed. © 2011 ACADEMY PUBLISHER.

关键词： Denial-of-service attack

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：