Czech Question Answering With Extended Sqad V3.0 Benchmark Dataset

RASLAN 2019: RECENT ADVANCES IN SLAVONIC NATURAL LANGUAGE PROCESSING(2019)

引用 0|浏览4
暂无评分
摘要
In this paper, we introduce a new version of the Simple Question Answering Databases (SQAD). The main asset of the new version lies in increasing the number of records to a total of 13,473 records. Besides the database enlargement, the new version incorporates new restrictions of specifying different formats of the expected answer for a given question. These new restrictions are connected with automatic database consistency checks where new sub-processes safeguard the database correctness and consistency.We also introduce a new on-line annotation tool used which offered a unified environment for extending the SQAD data in a crowdsourcing experiment.
更多
查看译文
关键词
question answering, QA benchmark dataset, SQAD, Czech
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要