賦權國小國語文學習:利用大型語言模型生成符合專業觀點的客製化教材
Empowering Elementary Learning: Utilizing Large Language Models to Craft Tailored Textbooks with Expert Insight
國立臺灣大學語言學研究所(Graduate Institute of Linguistics, National Taiwan University, Taipei, Taiwan)
摘要
大型語言模型(LLMs)近年憑藉其出色的零樣本(zero-shot)或少樣本(few-shot)學習能力,在各領域的研究中備受矚目,並成為許多專業工作流程中導入AI應用的關鍵技術。立基於深度語言學分析,本研究利用大型語言模型(GPT-4),針對臺灣國小學童自動生成客製化的國語文課文與生字練習。實驗初步結果顯示,模型不僅能產出符合指定年級程度的文本,其品質亦具有高度發展潛力。本研究的主要貢獻在於:首先,我們開創性地對臺灣現行國語文教科書進行量化分析;其次,我們設計了一套適用於不同學習程度的提示詞(prompts),成功利用大型語言模型為中文母語者自動生成教材。研究中亦包含模型生成內容與臺灣教育專家編寫版本之間的量化與質性比較分析。
Abstract
Large language models (LLMs) have in recent years spurred research across various sectors, owing to their remarkable zero-shot or few-shot performance. This capability has become indispensable for individuals seeking to integrate these language models into their workflows effectively. In this paper, based on in-depth linguistic analyses, we explore the application of an LLM, specifically GPT-4, in generating Chinese language textbooks tailored for grade school students. This encompasses the creation of main lesson texts alongside accompanying Chinese character exercises. Experimental results suggest that the LLM-generated textbook lessons are a viable research direction. The initial outcomes demonstrate the ability of LLM to generate texts of satisfactory quality appropriate for a specified grade level. The contributions of this work include pioneering the quantitative analysis of Chinese language textbooks for native speakers in Taiwan and leveraging an LLM to automatically generate textbook content and accompanying Chinese character exercises targeted at native Chinese speakers, which is a novel approach facilitated by the development of prompts tailored to different language learning levels. The study also conducts quantitative and qualitative comparisons between machine-generated lessons and those developed by educational professionals in Taiwan.
關鍵詞
- 類型
- 論文
- 期刊
- 圖書資訊學刊
- ISSN
- 1606-7509
- 卷期
- 23卷2期
- 出版
- 2025年12月
- 頁碼
- 145–183
- 編號
- NTUEJ00004468
同期目錄
- 腦波圖在網路搜尋研究中的應用:針對樣本大小、活躍作者和研究結果的系統性回顧Mahsa Torabi・頁 001–045
- 元宇宙數位策展之歷史事件體驗提升歷史學習成效陳志銘(Chih-Ming Chen);藍適齊(Shi-Chi Lan);李明娟(Ming-Chaun Li);翁榮秀(Jung-Hsiu Weng)・頁 047–083
- 疫情期間大學生由社群媒體接收錯誤訊息之資訊行為分析林麗娟(Lih-Juan ChanLin)・頁 085–118
- 應用類別對照之資料融合方法於機器學習研究分類黃建智(Chien-Chih Huang);陳光華(Kuang-Hua Chen)・頁 119–143
- 賦權國小國語文學習:利用大型語言模型生成符合專業觀點的客製化教材連大成(Da-Chen Lian);古貿昌(Mao-Chang Ku);王伯雅(Po-Ya Angela Wang);陳韋伶(Wei-Ling Chen);謝舒凱(Shu-Kai Hsieh)・頁 145–183
- 資料密集視角下的臺灣公民科學計畫:整體實踐與品質維護策略鄭瑋(Wei Jeng);甘婕人(Chieh-Jen Kan);黃寶霈(Pao-Pei Huang);陳靖瑄(Ching-Hsuan Chen)・頁 185–217
- 工作輪調對圖書館館員留任意願之影響:以工作特性與滿意度為中介變數廖子平(Tz-Ping Liao);柯皓仁(Hao-Ren Ke)・頁 219–257
- 蓮瓣辟蠹的文獻記載與實踐林振岳(Zhen-Yue Lin)・頁 259–282
- 《山東軍興紀略》作者及版本辨正王雨非(Yu-Fei Wang)・頁 283–307