TY - GEN
T1 - SPARC
T2 - 57th Annual Meeting of the Association for Computational Linguistics, ACL 2019
AU - Yu, Tao
AU - Zhang, Rui
AU - Yasunaga, Michihiro
AU - Tan, Yi Chern
AU - Lin, Xi Victoria
AU - Li, Suyi
AU - Er, Heyang
AU - Li, Irene
AU - Pang, Bo
AU - Chen, Tao
AU - Ji, Emily
AU - Dixit, Shreya
AU - Proctor, David
AU - Shim, Sungrok
AU - Kraft, Jonathan
AU - Zhang, Vincent
AU - Xiong, Caiming
AU - Socher, Richard
AU - Radev, Dragomir
N1 - Publisher Copyright:
© 2019 Association for Computational Linguistics
PY - 2020
Y1 - 2020
N2 - We present SParC, a dataset for cross-domain Semantic Parsing in Context. It consists of 4,298 coherent question sequences (12k+ individual questions annotated with SQL queries), obtained from controlled user interactions with 200 complex databases over 138 domains. We provide an in-depth analysis of SParC and show that it introduces new challenges compared to existing datasets. SParC (1) demonstrates complex contextual dependencies, (2) has greater semantic diversity, and (3) requires generalization to new domains due to its cross-domain nature and the unseen databases at test time. We experiment with two state-of-the-art text-to-SQL models adapted to the context-dependent, cross-domain setup. The best model obtains an exact set match accuracy of 20.2% over all questions and less than 10% over all interaction sequences, indicating that the cross-domain setting and the contextual phenomena of the dataset present significant challenges for future research. The dataset, baselines, and leaderboard are released at https://yale-lily.github.io/sparc.
AB - We present SParC, a dataset for cross-domain Semantic Parsing in Context. It consists of 4,298 coherent question sequences (12k+ individual questions annotated with SQL queries), obtained from controlled user interactions with 200 complex databases over 138 domains. We provide an in-depth analysis of SParC and show that it introduces new challenges compared to existing datasets. SParC (1) demonstrates complex contextual dependencies, (2) has greater semantic diversity, and (3) requires generalization to new domains due to its cross-domain nature and the unseen databases at test time. We experiment with two state-of-the-art text-to-SQL models adapted to the context-dependent, cross-domain setup. The best model obtains an exact set match accuracy of 20.2% over all questions and less than 10% over all interaction sequences, indicating that the cross-domain setting and the contextual phenomena of the dataset present significant challenges for future research. The dataset, baselines, and leaderboard are released at https://yale-lily.github.io/sparc.
UR - http://www.scopus.com/inward/record.url?scp=85084074849&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85084074849&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85084074849
T3 - ACL 2019 - 57th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference
SP - 4511
EP - 4523
BT - ACL 2019 - 57th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference
PB - Association for Computational Linguistics (ACL)
Y2 - 28 July 2019 through 2 August 2019
ER -