云南高校图书馆联盟文献共享服务平台- EXT

EXT: 作品数：362被引量：737H指数：14; 导出分析报告; 相关领域：自动化与计算机技术更多>>; 相关作者：梁英民韩骅黄斯勇张萍何飞更多>>; 相关机构：第四军医大学第四军医大学唐都医院西北师范大学安徽理工大学更多>>; 相关期刊：更多>>; 相关基金：国家自然科学基金国家高技术研究发展计划国家重点基础研究发展计划中国博士后科学基金更多>>

基于预训练模型与双向注意力流的抽取式阅读理解模型: 《首都师范大学学报(自然科学版)》2025年第2期1-11,共11页文勇军吴金铭梅硕; 针对服务机器人的抽取式阅读理解任务中出现答案抽取准确度不高的问题,构建了基于预训练模型与双向注意力流的抽取式阅读理解模型。该模型首先采用预训练模型来提取问题与文档上下文的浅层联合语义表征;其次利用双向注意力网络来加强特...; 关键词：自然语言处理机器阅读理解预训练模型双向注意力流(BERT) RoBERTa-wwm-ext 答案抽取

基于DP-Textrank的手机产品消费者需求偏好研究: 《经营与管理》2025年第3期61-67,共7页钟正徐子鸣黄逸超; 随着科技水平的提高和物质生活的丰富,消费者对产品的需求和期望也越来越高。能否针对不同消费层次消费者产出最能迎合消费者属性需求的商品线,成为企业能否成功的关键。随着电商平台的发展,消费者为了分享购物体验发表了许多在线评论信...; 关键词：在线评论需求偏好依存句法分析文本摘要

Differential Privacy-Enabled TextCNN for MOOCs Fake Review Detection: 《Journal of Electronic Research and Application》2025年第1期191-201,共11页Caiyun Chen; The rapid development and widespread adoption of massive open online courses(MOOCs)have indeed had a significant impact on China’s education curriculum.However,the problem of fake reviews and ratings on the platform ...; 关键词：DP-TextCNN Differential Privacy Fake review MOOCs

融入夸张表征的中文反讽识别方法: 《数据分析与知识发现》2025年第2期1-11,共11页李书羽朱广丽李嘉伟段文杰周若彤张顺香; 国家自然科学基金面上项目(项目编号:62076006);认知智能全国重点实验室开放课题(项目编号:COGOS-2023HE02);安徽高校协同创新项目(项目编号:GXXT-2021-008)的研究成果之一。; 【目的】为解决中文反讽短文本中存在的特征稀疏问题,提出一种融入夸张表征的中文反讽识别方法,挖掘短文本中的夸张表征以提升中文反讽识别准确率。【方法】通过点互信息和语义相似度计算分别获取与反讽领域相关的共现词对集、感叹词集...; 关键词：中文反讽领域词典夸张表征 RoBERTa-wwm-ext 多头注意力机制

Skid resistance performance and texture lateral distribution within the lanes of asphalt pavements: 《Journal of Traffic and Transportation Engineering(English Edition)》2025年第1期87-107,共21页Di Yun Liqun Hu Ulf Sandberg Cheng Tang; supported by National Key R&D Program of China(No.2018YFB1600200);the Fundamental Research Funds for the Central Universities(Nos.310821173101,300102218515)。; The skid resistance and pavement texture can vary a lot for different lane paths,meaning that the lateral shift of the vehicle driving in the lane section can affect the safety significantly.On the other hand,a varyin...; 关键词：Lane section Skid resistance Pavement texture Drainage Lateral position

Relationship Between Stress and Texture in L1_(0)-FePt Thin Films: 《稀有金属材料与工程》2025年第2期337-342,共6页Wang Xuanli Li Wei; Inner Mongolia Natural Science Foundation Project(2020LH05028)。; Impact of texture type on the magnetic properties of ultrahigh density perpendicular magnetic recording media L1_(0)-FePt thin film was investigated,so were the texture formation and evolution mechanism.Reuss,Voigt,an...; 关键词：L1_(0)-FePt film TEXTURE STRESS elastic modulus

Dr.ICL:Demonstration-Retrieved In-context Learning: 《Data Intelligence》2024年第4期909-922,共14页Man Luo Xin Xu Zhuyun Dai Panupong Pasupat Mehran Kazemi Chitta Baral Vaiva Imbrasaite Vincent Y Zhao; In-context learning(ICL), which teaches a large language model(LLM) to perform a task with few-shot demonstrations rather than adjusting the model parameters, has emerged as a strong paradigm for using LLMs. While ear...; 关键词：Information retrieval In-context learning Large language models Retrieval augmented generation Large language model reasoning

Teaching Life Writing: Theory and Textual Practice: 《Cultural and Religious Studies》2024年第12期741-746,共6页Tzu Yu Allison Lin Erhan Yokuş; In this research about teaching life-writing,the authors would like to focus on different ways of demonstrating narratives about personal lives.Some established research articles and academic writings on autobiography...; 关键词：WRITING NARRATIVE auto/biography TEACHING CLASSROOM

Image Enhancement via Associated Perturbation Removal and Texture Reconstruction Learning: 《IEEE/CAA Journal of Automatica Sinica》2024年第11期2253-2269,共17页Kui Jiang Ruoxi Wang Yi Xiao Junjun Jiang Xin Xu Tao Lu; supported by the National Natural Science Foundation of China (U23B2009, 62376201, 423B2104);Open Foundation (ZNXX2023MSO2, HBIR202311)。; Degradation under challenging conditions such as rain, haze, and low light not only diminishes content visibility, but also results in additional degradation side effects, including detail occlusion and color distorti...; 关键词：Association learning attention mechanism image enhancement perturbation modeling

Ensemble Filter-Wrapper Text Feature Selection Methods for Text Classification: 《Computer Modeling in Engineering & Sciences》2024年第11期1847-1865,共19页Oluwaseun Peter Ige Keng Hoon Gan; supported by Universiti Sains Malaysia(USM)and School of Computer Sciences,USM。; Feature selection is a crucial technique in text classification for improving the efficiency and effectiveness of classifiers or machine learning techniques by reducing the dataset’s dimensionality.This involves elim...; 关键词：Metaheuristic algorithms text classification multi-univariate filter feature selection ensemble filter-wrapper techniques

EXT