跳至主導覽 跳至搜尋 跳過主要內容

Fine-Tuning and Evaluation of Question Generation for Slovak Language

  • Ondrej Megela
  • , Daniel Hládek
  • , Matúš Pleva
  • , Ján Staš
  • , Ming Hsiang Su
  • , Yuan Fu Liao

研究成果: Conference contribution同行評審

1 引文 斯高帕斯(Scopus)

摘要

Automatic generation of questions about the given context is useful for the adaptation of question-answering systems or to support education. We trained and evaluated a model that generates a question in the Slovak language. We have designed an automatic metric where an additional question-answering model is used to evaluate the generated questions. We calculated how many questions have confidence greater than the given threshold. For generating questions, we used contexts from the Slovak question-answering dataset. The fine-tuned Slovak T5 model did generate 38% of the questions that the evaluation model could answer with confidence greater than 50%. We cooperated with partners from Taiwan during these experiments in the frame of a bilateral project and we plan to transfer the knowledge to the Chinese language later.

原文English
主出版物標題ROCLING 2023 - Proceedings of the 35th Conference on Computational Linguistics and Speech Processing
編輯Jheng-Long Wu, Ming-Hsiang Su, Hen-Hsen Huang, Yu Tsao, Hou-Chiang Tseng, Chia-Hui Chang, Lung-Hao Lee, Yuan-Fu Liao, Wei-Yun Ma
發行者The Association for Computational Linguistics and Chinese Language Processing (ACLCLP)
頁面171-178
頁數8
ISBN(電子)9789869576963
出版狀態Published - 2023
事件35th Conference on Computational Linguistics and Speech Processing, ROCLING 2023 - Taipei City, 台灣
持續時間: 20 10月 202321 10月 2023

出版系列

名字ROCLING 2023 - Proceedings of the 35th Conference on Computational Linguistics and Speech Processing

Conference

Conference35th Conference on Computational Linguistics and Speech Processing, ROCLING 2023
國家/地區台灣
城市Taipei City
期間20/10/2321/10/23

指紋

深入研究「Fine-Tuning and Evaluation of Question Generation for Slovak Language」主題。共同形成了獨特的指紋。

引用此