检索结果-内蒙古大学图书馆

A Survey of LLM datasets:From Autoregressive Model to AI Chatbot

Journal of Computer science & technology 2024年第3期39卷 542-566页

作者：杜非马新建杨婧如柳熠罗超然王学斌姜海鸥景翔 National Key Laboratory of Data Space Technology and System Beijing 100195China Advanced Institute of Big Data Beijing 100195China Fu Foundation School of Engineering and Applied Science Columbia UniversityNY 10027U.S.A. School of Software and Microelectronics Peking UniversityBeijing 100091China CCF IEEE

Since OpenAI opened access to ChatGPT,large language models(LLMs)become an increasingly popular topic attracting researchers’attention from abundant ***,public researchers meet some problems when developing LLMs given that most of the LLMs are produced by industries and the training details are typically *** datasets are an important setup of LLMs,this paper does a holistic survey on the training datasets used in both the pre-train and fine-tune *** paper first summarizes 16 pre-train datasets and 16 fine-tune datasets used in the state-of-the-art ***,based on the properties of the pre-train and fine-tune processes,it comments on pre-train datasets from quality,quantity,and relation with models,and comments on fine-tune datasets from quality,quantity,and *** study then critically figures out the problems and research trends that exist in current LLM *** study helps public researchers train and investigate LLMs by visual cases and provides useful comments to the research community regarding data *** the best of our knowledge,this paper is the first to summarize and discuss datasets used in both autoregressive and chat *** survey offers insights and suggestions to researchers and LLM developers as they build their models,and contributes to the LLM study by pointing out the existing problems of LLM studies from the perspective of data.

关键词： large language model(LLM) autoregressive model AI chatbot natural language processing(NLP)corpora OpenAI

来源：评论

学校读者我要写书评

暂无评论

An End-to-end Learning Based Covolutional Neural Network for Single Image Defogging Algorithm 2023

An End-to-end Learning Based Covolutional Neural Network for...

引用

5th International Symposium on Signal Processing Systems, SSPS 2023

作者： Li, Qiqing Li, Ru Shen, Xin Lv, Wei School of Aliyun Big Data Applications Zhuhai College of Science and Technology China School of Aliyun Big Data Applications City University of Macau China

ISBN: (纸本)9798400700040

In the era of big data, there are more and more outdoor camera acquisition equipment. Due to the influence of extreme weather, such as fog, camera acquisition equipment is easy to lead to the decline of image quality and destroy the value of image application. Therefore, this paper will propose an advanced dehazing algorithm to make the foggy image clearer. Based on the principle of residual neural network, combined with attention mechanism and feature pyramid idea, this paper proposes an end to-end learning single image dehazing algorithm. Let the network learn the relationship between channels and pixels, and use the feature pyramid multi-scale fusion feature to restore the foggy image to a clear image. The SSIM score was 0.9687 and the PSNR score was 29.16. Very good results have been achieved on the RESIDE outdoor dataset. This paper finds the scores obtained by testing DCP, AOD-NET, DeHazeNet, and GFN methods on the same dataset. Compared with these four methods, there is a significant improvement. In particular, it is 15.39% higher than the DCP method on SSIM and 10.03% higher on PSNP. © 2023 ACM.

关键词： Statistical tests

来源：评论

学校读者我要写书评

暂无评论

A Diversion Path Planning Method of Waypoint Based on Flight Traffic Conduction 6

A Diversion Path Planning Method of Waypoint Based on Flight...

引用

6th IEEE International Conference on Civil Aviation Safety and Information technology, ICCASIT 2024

作者： Shi, Hongfang Le, Ningning Miao, Jiahe Li, Qian China Academy of Civil Aviation Science and Technology Big Data Analysis and Application Center Beijing China

ISBN: (纸本)9798350389418

Based on flight operation data, this paper constructs a diversion path planning method for busy waypoints by analyzing the relationship of flight traffic conduction between waypoints. Taking busy waypoint KHN as an example, a total of two diversion paths are planned, specifically one in the east-west direction and one in the north-south direction, and the two diversion paths are evaluated in terms of safety and economy, and the results of the evaluations are within the acceptable range. The method of this paper can improve the capacity of airspace resources without changing the existing air route network;at the same time, this method can also assist in identifying the bottleneck of airspace resources, which means identifying the key waypoints affecting the flow of diversion paths, so as to provide certain theoretical references for China to enhance the total number of flights, and to improve the utilization rate of airspace resources and operational efficiency. © 2024 IEEE.

关键词： Flight paths

来源：评论

学校读者我要写书评

暂无评论

JobViz:Skill-driven visual exploration of job advertisements

引用

Visual Informatics 2024年第3期8卷 18-28页

作者： Ran Wang Qianhe Chen Yong Wang Lewei Xiong Boyang Shen School of Journalism and Information Communication Huazhong University of Science and TechnologyWuhanChina School of Computer Science and Technology Huazhong University of Science and TechnologyWuhanChina College of Computing and Data Science Nanyang Technological UniversitySingaporeSingapore Wuhan National Laboratory for Optoelectronics WuhanChina Philosophy and Social Science Laboratory of Big Data and National Communication Strategy Ministry of EducationWuhanChina

Online job advertisements on various job portals or websites have become the most popular way for people to find potential career opportunities ***,the majority of these job sites are limited to offering fundamental filters such as job titles,keywords,and compensation *** often poses a challenge for job seekers in efficiently identifying relevant job advertisements that align with their unique skill sets amidst a vast sea of ***,we propose well-coordinated visualizations to provide job seekers with three levels of details of job information:a skill-job overview visualizes skill sets,employment posts as well as relationships between them with a hierarchical visualization design;a post exploration view leverages an augmented radar-chart glyph to represent job posts and further facilitates users’swift comprehension of the pertinent skills necessitated by respective positions;a post detail view lists the specifics of selected job posts for profound analysis and *** using a real-world recruitment advertisement dataset collected from 51Job,one of the largest job websites in China,we conducted two case studies and user interviews to evaluate *** results demonstrated the usefulness and effectiveness of our approach.

关键词： Visual exploration Job advertisements Skill-driven

来源：评论

学校读者我要写书评

暂无评论

Towards High-Performance Graph Processing: From a Hardware/Software Co-Design Perspective

引用

Journal of Computer science & technology 2024年第2期39卷 245-266页

作者：廖小飞赵文举金海姚鹏程黄禹王庆刚赵进郑龙张宇邵志远 National Engineering Research Center for Big Data Technology and System School of Computer Science and Technology Huazhong University of Science and TechnologyWuhan 430074China Services Computing Technology and System Laboratory School of Computer Science and Technology Huazhong University of Science and TechnologyWuhan 430074China Cluster and Grid Computing Laboratory School of Computer Science and TechnologyHuazhong University of Science and TechnologyWuhan 430074China Zhejiang Lab Hangzhou 311121China

Graph processing has been widely used in many scenarios,from scientific computing to artificial *** processing exhibits irregular computational parallelism and random memory accesses,unlike traditional ***,running graph processing workloads on conventional architectures(e.g.,CPUs and GPUs)often shows a significantly low compute-memory ratio with few performance benefits,which can be,in many cases,even slower than a specialized single-thread graph *** domain-specific hardware designs are essential for graph processing,it is still challenging to transform the hardware capability to performance boost without coupled software *** article presents a graph processing ecosystem from hardware to *** start by introducing a series of hardware accelerators as the foundation of this ***,the codesigned parallel graph systems and their distributed techniques are presented to support graph ***,we introduce our efforts on novel graph applications and hardware *** results show that various graph applications can be efficiently accelerated in this graph processing ecosystem.

关键词： graph processing hardware accelerator software system high performance ecosystem

来源：评论

学校读者我要写书评

暂无评论

Multi-Dimensional Training Optimization for Efficient Federated Synergy Learning

引用

IEEE Transactions on Mobile Computing 2025年第7期24卷 6243-6258页

作者： Fu, Shucun Dong, Fang Chen, Runze Shen, Dian Zhang, Jinghui He, Qiang Southeast University School of Computer Science and Engineering Nanjing 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 Wuhan China

Edge learning (EL) is an end-to-edge collaborative learning paradigm enabling devices to participate in model training and data analysis, opening countless opportunities for edge intelligence. As a promising EL framework, federated synergy learning (FSyL) mitigates the computation and communication overhead on resource-constrained devices by offloading partial model layers to the edge server for synergistic training. Nevertheless, due to the system and statistical heterogeneity, naively using existing FSyL methods is significantly time-consuming and causes accuracy degradation. Motivated by this issue, this paper introduces a novel FSyL framework that integrates multi-dimensional training optimization and formulates the edge learning cost minimization (ELCM) problem. To tackle the ELCM efficiently, we design OL-MG, an OnLine Model Splitting and Resource Provisioning Game. Specifically, we first reformulate and decompose the original ELCM based on data quality evaluation. Then, given a model splitting decision, we determine the optimal resource provisioning in Sub-problem1, based on which optimal model splitting in Sub-problem2 is modeled as a potential game. Subsequently, we introduce a decentralized algorithm to find a Nash equilibrium (NE) solution. Furthermore, we further extend OL-MG to support a budget-aware multi-edge scenario. Extensive experiments demonstrate that the proposed mechanism significantly outperforms state-of-the-art methods in cost-saving and accuracy improvement. © 2025 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

An efficient Quasi-Affine Transformation Evolutionary algorithm with fixed dimension updating and its application in UAV 3D path planning

引用

Journal of Intelligent and Fuzzy Systems 2024年第4期46卷 9755-9781页

作者： Sung, Tien-Wen Zhao, Baohua Zhang, Xin Lee, Chao-Yang Fang, Qingjun Fujian Provincial Key Laboratory of Big Data Mining and Applications College of Computer Science and Mathematics Fujian University of Technology Fuzhou China Department of Computer Science and Information Engineering National Yunlin University of Science and Technology Yunlin Taiwan

Quasi-Affine Transformation Evolutionary (QUATRE) algorithm is a kind of swarm-based collaborative optimization algorithm that solves the problem of a position deviation in a DE search by using the co-evolution matrix M instead of the cross-control parameter CR in the differential evolution algorithm (DE). However, QUATRE shares some of the same weaknesses as DE, such as premature convergence and search stagnation. Inspired by the artificial bee colony algorithm (ABC), we propose a new QUATRE algorithm to improve these problems that ranks all the individuals and evolves only the poorer half of the population. In an evolving population, individuals of different levels intersect with dimensions of different sizes to improve search efficiency and accuracy. In addition, we establish a better selection framework for the parent generation individuals and select more excellent parent individuals to complete the evolution for the individuals trapped in search stagnation. To verify the performance of the new QUATRE algorithm, we divide the comparison algorithm into three groups, including ABC variant group, DE variant group, and QUATRE variant group, and the CEC2014 test suite is used for the comparison. The experimental results show the new QUATRE algorithm performance is competitive. We also successfully apply the new QUATRE algorithm on the 3D path planning of UAV, and compared with the other famous algorithm performance it is still outstanding, which verifies the algorithm's practicability. © 2024 - IOS Press. All rights reserved.

关键词： Optimization

来源：评论

学校读者我要写书评

暂无评论

Finite-Time Anti-Synchronization for Memristive Neural Networks with Time-Varying Delays

引用

IAENG International Journal of Applied Mathematics 2025年第3期55卷 611-617页

作者： Duan, Lian Zhang, Ziyue Li, Ziyang Anhui Province Engineering Laboratory for Big Data Analysis and Early Warning Technology of Coal Mine Safety Department of Mathematics Anhui University of Science and Technology Huainan232001 China Department of Mathematics Anhui University of Science and Technology Huainan232001 China

This paper addresses the finite-time anti-synchronization issue for a type of delayed memristive neural networks. By designing a novel memoryless state-feedback controller, novel criteria on finite-time anti-synchronization of the addressed system are discovered based on drive-response framework, rigorous mathematical analysis techniques and differential inclusions theory. The established theoretical results indicate that the switching between finite-time and fixed-time anti-synchronization depends on the position of the initial functions, which are essentially different from existing switching mechanism. In addition, a simulated example is given to verify the validity of the theoretical findings. © (2025), (International Association of Engineers). All rights reserved.

关键词： State feedback

来源：评论

学校读者我要写书评

暂无评论

A Reduced State-Space Generation Method for Concurrent Systems Based on CPN-PR Model

引用

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2025年第6期44卷 2328-2342页

作者： Zhong, Wenjie Sun, Tao Zhou, Jian-Tao Wang, Zhuowei Song, Xiaoyu Inner Mongolia University College of Computer Science the Engineering Research Center of Ecological Big Data Ministry of Education the Inner Mongolia Engineering Laboratory for Cloud Computing and Service Software the Inner Mongolia Engineering Laboratory for Big Data Analysis Technology Hohhot010000 China Guangdong University of Technology School of Computer Science and Technology Guangzhou510006 China Portland State University Department of Electrical and Computer Engineering PortlandOR97207 United States

Colored Petri nets (CPNs) provide descriptions of the concurrent behaviors for software and hardware. Model checking based on CPNs is an effective method to simulate and verify the concurrent behavior in system design. However, the model-checking method traverses the full state space, which suffers from the state-space explosion problem. A reduced state-space generation method related to the property of concurrent systems is proposed. Specifically, we extend CPNs to define a property-related model (CPN-PR) and give a property-related analysis method whose results can be used to generate the CPN-PR model. A reduced state-space generation method is developed based on enabled binding element filtering rules. The stutter trace equivalence between the state spaces of CPN and CPN-PR has been proven by showing that the reduced state space may not change the model-checking result. A comparison experiment is conducted to demonstrate the effectiveness of our method. © 1982-2012 IEEE.

关键词： Model checking

来源：评论

学校读者我要写书评

暂无评论

BGNN: Behavior-aware graph neural network for heterogeneous session-based recommendation

引用

Frontiers of Computer science 2023年第5期17卷 103-118页

作者： Jinwei LUO Mingkai HE Weike PAN Zhong MING College of Computer Science and Software Engineering Shenzhen UniversityShenzhen 518060China National Engineering Laboratory for Big Data System Computing Technology Shenzhen UniversityShenzhen 518060China

Session-based recommendation(SBR)and multibehavior recommendation(MBR)are both important problems and have attracted the attention of many researchers and *** from SBR that solely uses one single type of behavior sequences and MBR that neglects sequential dynamics,heterogeneous SBR(HSBR)that exploits different types of behavioral information(e.g.,examinations like clicks or browses,purchases,adds-to-carts and adds-to-favorites)in sequences is more consistent with real-world recommendation scenarios,but it is rarely *** efforts towards HSBR focus on distinguishing different types of behaviors or exploiting homogeneous behavior transitions in a sequence with the same type of ***,all the existing solutions for HSBR do not exploit the rich heterogeneous behavior transitions in an explicit way and thus may fail to capture the semantic relations between different types of ***,all the existing solutions for HSBR do not model the rich heterogeneous behavior transitions in the form of graphs and thus may fail to capture the semantic relations between different types of *** limitation hinders the development of HSBR and results in unsatisfactory *** a response,we propose a novel behavior-aware graph neural network(BGNN)for *** BGNN adopts a dual-channel learning strategy for differentiated modeling of two different types of behavior sequences in a ***,our BGNN integrates the information of both homogeneous behavior transitions and heterogeneous behavior transitions in a unified *** then conduct extensive empirical studies on three real-world datasets,and find that our BGNN outperforms the best baseline by 21.87%,18.49%,and 37.16%on average correspondingly.A series of further experiments and visualization studies demonstrate the rationality and effectiveness of our *** exploratory study on extending our BGNN to handle more than two types of behaviors show that our BGNN can e

关键词： session-based recommendation graph neural network heterogeneous behaviors

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：