Abstract
We propose Grounded Adaptation for Zeroshot Executable Semantic Parsing (GAZP) to adapt an existing semantic parser to new environments (e.g. new database schemas). GAZP combines a forward semantic parser with a backward utterance generator to synthesize data (e.g. utterances and SQL queries) in the new environment, then selects cycle-consistent examples to adapt the parser. Unlike data-augmentation, which typically synthesizes unverified examples in the training environment, GAZP synthesizes examples in the new environment whose input-output consistency are verified through execution. On the Spider, Sparc, and CoSQL zero-shot semantic parsing tasks, GAZP improves logical form and execution accuracy of the baseline parser. Our analyses show that GAZP outperforms data-augmentation in the training environment, performance increases with the amount of GAZP-synthesized data, and cycle-consistency is central to successful adaptation.- Anthology ID:
- 2020.emnlp-main.558
- Volume:
- Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
- Month:
- November
- Year:
- 2020
- Address:
- Online
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 6869–6882
- Language:
- URL:
- https://aclanthology.org/2020.emnlp-main.558
- DOI:
- 10.18653/v1/2020.emnlp-main.558
- Cite (ACL):
- Victor Zhong, Mike Lewis, Sida I. Wang, and Luke Zettlemoyer. 2020. Grounded Adaptation for Zero-shot Executable Semantic Parsing. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6869–6882, Online. Association for Computational Linguistics.
- Cite (Informal):
- Grounded Adaptation for Zero-shot Executable Semantic Parsing (Zhong et al., EMNLP 2020)
- PDF:
- https://preview.aclanthology.org/nodalida-main-page/2020.emnlp-main.558.pdf
- Data
- CoSQL, SParC