Multimodal pretraining has made convincing achievements in various downstream tasks in recent years.However,since the majority of the existing works construct models based on English,their applications are limited by ...Multimodal pretraining has made convincing achievements in various downstream tasks in recent years.However,since the majority of the existing works construct models based on English,their applications are limited by language.In this work,we address this issue by developing models with multimodal and multilingual capabilities.We explore two types of methods to extend multimodal pretraining model from monolingual to multilingual.Specifically,we propose a pretraining-based model named multilingual multimodal pretraining(MLMM),and two generalization-based models named multilingual CLIP(M-CLIP)and multilingual acquisition(MLA).In addition,we further extend the generalization-based models to incorporate the audio modality and develop the multilingual CLIP for vision,language,and audio(CLIP4VLA).Our models achieve state-of-the-art performances on multilingual vision-text retrieval,visual question answering,and image captioning benchmarks.Based on the experimental results,we discuss the pros and cons of the two types of models and their potential practical applications.展开更多
针对维吾尔文字、哈萨克文字、柯尔克孜文字(以下简称“维哈柯文”)的特点以及进行维哈柯文、西文等多语种混合处理时的特殊需求,本文通过对L inux的I18N体系中NLS(National Language Support)研究分析,提出了基于L inux的多语种图形化...针对维吾尔文字、哈萨克文字、柯尔克孜文字(以下简称“维哈柯文”)的特点以及进行维哈柯文、西文等多语种混合处理时的特殊需求,本文通过对L inux的I18N体系中NLS(National Language Support)研究分析,提出了基于L inux的多语种图形化处理平台的设计目标与总体架构。该平台由维哈柯文本地化环境、维哈柯文显示、自适应维哈柯文输入和维哈柯文打印输出等4个子系统的十余个模块组成。本文详细介绍了各子系统主要模块的实现技术。通过在redhat linux 8.0、turbolinux上测试表明,该平台在桌面环境、编辑软件、网络浏览、数据库软件、多媒体软件、图形处理软件等应用中均能较好的实现维哈柯文、汉文、西文的混合输入、显示、编辑、排版、打印等功能。展开更多
Purpose This study aimed to explore how educational technology influenced a preparatory teacher education program using heteroglossic and global approaches.Design/Approach/Methods The researchers drew upon the theoret...Purpose This study aimed to explore how educational technology influenced a preparatory teacher education program using heteroglossic and global approaches.Design/Approach/Methods The researchers drew upon the theoretical framework of multilingual digital storytelling(MDST),which emphasizes the intercultural awareness attributes of multilingual learners(MLs)and takes a heteroglossic perspective in linguistic pedagogy.This qualitative case study examined the experiences of 11 U.S.-based preservice teachers(PSTs)and 12 MLs elementary students in Korea in the MDST project of a TESOL methods course.Findings The findings showed that PSTs and MLs enhanced their appreciation for educational technology,multilingualism,and intercultural awareness.Although the project aimed to decenter English as a hegemonic language,both PSTs and MLs maintained traditional discourses that privileged English over MLs’home language and targeted literacy correctness in written English only,moving away from the heterogeneous goals of the course project.PSTs and MLs also faced challenges in navigating technological tools,which negatively affected their perception of the project.Originality/Value This study contributes to heteroglossic approaches in preparatory TESOL teacher education programs and improves the understanding of challenges in educational technology use for global multilingual exchanges to promote global citizenship.展开更多
文章在前人研究的基础上,结合藏文文献,对柏林勃兰登堡科学院藏吐鲁番出土的TibHT 107号(T II D 19和T II D 20)梵藏回鹘文多语种《二十一度母礼赞经》文本和版画残片进行了再次研究。结果显示,其中的藏文赞词来源于《甘珠尔》同名经典...文章在前人研究的基础上,结合藏文文献,对柏林勃兰登堡科学院藏吐鲁番出土的TibHT 107号(T II D 19和T II D 20)梵藏回鹘文多语种《二十一度母礼赞经》文本和版画残片进行了再次研究。结果显示,其中的藏文赞词来源于《甘珠尔》同名经典,其中T II D 19号残片中的藏文为第六位度母释梵火天母的赞词,T II D 20号残片中的藏文为第五位度母怛啰吽字母的赞词。吐鲁番版上图下文的版式受到内地唐末以来传统的影响;经与文献比对,版画中度母的造型按日护传规的样式创作而成,绘画风格体现出浓郁的元代宫廷藏传艺术特点;多语种文本和版画综合特征显示,吐鲁番版极有可能来自元朝首都北京或按其刻本仿制而成。该残片为元代多民族文化交融的具体结晶。展开更多
基金supported by the National Natural Science Foundation of China(No.62072462)the National Key R&D Program of China(No.2020AAA0108600)the Large-scale Pretraining Program 468 of Beijing Academy of Artificial Intelligence(BAAI).
文摘Multimodal pretraining has made convincing achievements in various downstream tasks in recent years.However,since the majority of the existing works construct models based on English,their applications are limited by language.In this work,we address this issue by developing models with multimodal and multilingual capabilities.We explore two types of methods to extend multimodal pretraining model from monolingual to multilingual.Specifically,we propose a pretraining-based model named multilingual multimodal pretraining(MLMM),and two generalization-based models named multilingual CLIP(M-CLIP)and multilingual acquisition(MLA).In addition,we further extend the generalization-based models to incorporate the audio modality and develop the multilingual CLIP for vision,language,and audio(CLIP4VLA).Our models achieve state-of-the-art performances on multilingual vision-text retrieval,visual question answering,and image captioning benchmarks.Based on the experimental results,we discuss the pros and cons of the two types of models and their potential practical applications.
文摘针对维吾尔文字、哈萨克文字、柯尔克孜文字(以下简称“维哈柯文”)的特点以及进行维哈柯文、西文等多语种混合处理时的特殊需求,本文通过对L inux的I18N体系中NLS(National Language Support)研究分析,提出了基于L inux的多语种图形化处理平台的设计目标与总体架构。该平台由维哈柯文本地化环境、维哈柯文显示、自适应维哈柯文输入和维哈柯文打印输出等4个子系统的十余个模块组成。本文详细介绍了各子系统主要模块的实现技术。通过在redhat linux 8.0、turbolinux上测试表明,该平台在桌面环境、编辑软件、网络浏览、数据库软件、多媒体软件、图形处理软件等应用中均能较好的实现维哈柯文、汉文、西文的混合输入、显示、编辑、排版、打印等功能。
文摘Purpose This study aimed to explore how educational technology influenced a preparatory teacher education program using heteroglossic and global approaches.Design/Approach/Methods The researchers drew upon the theoretical framework of multilingual digital storytelling(MDST),which emphasizes the intercultural awareness attributes of multilingual learners(MLs)and takes a heteroglossic perspective in linguistic pedagogy.This qualitative case study examined the experiences of 11 U.S.-based preservice teachers(PSTs)and 12 MLs elementary students in Korea in the MDST project of a TESOL methods course.Findings The findings showed that PSTs and MLs enhanced their appreciation for educational technology,multilingualism,and intercultural awareness.Although the project aimed to decenter English as a hegemonic language,both PSTs and MLs maintained traditional discourses that privileged English over MLs’home language and targeted literacy correctness in written English only,moving away from the heterogeneous goals of the course project.PSTs and MLs also faced challenges in navigating technological tools,which negatively affected their perception of the project.Originality/Value This study contributes to heteroglossic approaches in preparatory TESOL teacher education programs and improves the understanding of challenges in educational technology use for global multilingual exchanges to promote global citizenship.
文摘文章在前人研究的基础上,结合藏文文献,对柏林勃兰登堡科学院藏吐鲁番出土的TibHT 107号(T II D 19和T II D 20)梵藏回鹘文多语种《二十一度母礼赞经》文本和版画残片进行了再次研究。结果显示,其中的藏文赞词来源于《甘珠尔》同名经典,其中T II D 19号残片中的藏文为第六位度母释梵火天母的赞词,T II D 20号残片中的藏文为第五位度母怛啰吽字母的赞词。吐鲁番版上图下文的版式受到内地唐末以来传统的影响;经与文献比对,版画中度母的造型按日护传规的样式创作而成,绘画风格体现出浓郁的元代宫廷藏传艺术特点;多语种文本和版画综合特征显示,吐鲁番版极有可能来自元朝首都北京或按其刻本仿制而成。该残片为元代多民族文化交融的具体结晶。
