license: cc
language:
- en
- fr
- de
- vi
- zh
- ja
- es
size_categories:
- 10K<n<100K
MultiSpider: Towards Benchmarking Multilingual Text-to-SQL Semantic Parsing
In this work, we present MultiSpider, a multilingual text-to-SQL dataset which covers seven languages (English, German, French, Spanish, Japanese, Chinese, and Vietnamese). Find more details on paper and code.
Please be aware that the MultiSpider dataset is available in two versions: with_English_value and with_original_value. Our reported results are based on the with_English_value version to circumvent any discrepancies between the entities in the questions and the values in the database.
The with_original_value version is a byproduct of the dataset creation process, which may be of interest for more in-depth research on this localized dataset.
with_English_value: Führen Sie die Namen der Sängerinnen und Sänger auf, deren Staatsbürgerschaft nicht „France“ lautet.
with_original_value: Führen Sie die Namen der Sängerinnen und Sänger auf, deren Staatsbürgerschaft nicht "Frankreich" lautet.
Results
| Model | EN | DE | ES | FR | JA | ZH | VI |
|---|---|---|---|---|---|---|---|
| Paper Report | 68.8 | 64.8 | 67.4 | 65.3 | 60.2 | 66.1 | 67.1 |
| Released Model | 69.5 | 65.1 | 68.1 | 66.7 | 60.9 | 67.4 | 69.1 |
Copyright
Except where stated explicitly otherwise, the copyright to the source code is licensed under the Creative Commons - Attribution-NonCommercial 4.0 International license (CC BY-NC 4.0): https://creativecommons.org/licenses/by-nc/4.0/.
Any commercial use (whether for the benefit of third parties or internally in production) requires an explicit license.