TY - DATA T1 - SpiderDec, the decomposed version of the Spider dev data set PY - 2024/07/01 AU - Sara Salimzadeh AU - Ujwal Gadiraju AU - Claudia Hauff AU - Arie Van Deursen UR - DO - 10.4121/66ab9ab1-a08a-4c63-bd92-d11e2c3c06f8.v1 KW - Text-to-SQL KW - Semantic Parsing KW - Natural Language Interface to Databases KW - Corpus Annotation N2 -
SpiderDec is an extension of the Spider Dataset. The original Spider dataset split the data into training, development, and a hidden test set. For this new dataset, we manually decomposed the questions and corresponding queries within the development set of the Spider dataset, focusing on those with hard and extra hard SQL queries. The result of this effort is the creation of SpiderDec.
ER -