大语言模型在二语教学中的应用效能解析  

Efficacy assessment of large language models application in L2 teaching

作　　者：苏祺 SU Qi

机构地区：[1]北京大学外国语学院外国语言学及应用语言学研究所、人工智能研究院、数字人文研究中心,北京100871

出　　处：《外语界》2024年第3期35-42,共8页Foreign Language World

基　　金：“中央高校基本科研业务费”资助。

摘　　要：本研究聚焦大语言模型在二语教学的语法纠错、语言翻译、智能写作与写作评分、智能语伴与口语评估等重要领域的应用,通过对比大语言模型与其他语言智能系统在特定语言学习场景下的表现,探讨大语言模型的应用前景及其对二语学习成效的影响。文章强调了大语言模型在提供实时反馈、个性化学习体验、多语言支持等方面的优势,也指出了其在语境理解等方面的缺陷,以期明确大语言模型在二语教学中的适用性与局限性,为二语教学中的人工智能工具选择应用和教育技术发展提供有益参考。This study focuses on the application of large language models(LLMs)in key areas of L2 teaching,including grammatical error correction,language translation,AI-assisted writing and scoring,and AI-assisted spoken language learning and assessment.By comparing the performance of LLMs with specialized AI systems in specific L2 teaching scenarios,the study explores the prospects of LLMs application and its impact on the effectiveness of L2 learning.It highlights the strengths of LLMs in providing real-time feedback,personalized learning experiences,and multi-language support,while also acknowledging their limitations,such as challenges in context understanding.The study aims to clarify the applicability and limitations of LLMs in L2 teaching,and provide useful reference for the selection and adoption of AI tools in L2 teaching and development of educational technology.

关 键 词：大语言模型 语言智能 二语教学 应用效能 

分 类 号：H319[语言文字—英语]