Grammar-based Neural Text-to-SQL Generation
By Kevin Lin, Ben Bogin, Mark Neumann, Jonathan Berant, Matt Gardner. 2019
65 citations recorded by the source corpus.
The sequence-to-sequence paradigm employed by neural text-to-SQL models typically performs token-level decoding and does not consider generating SQL hierarchically from a grammar. Grammar-based decoding has shown significant improvements for other semantic parsing tasks, but SQL and other general programming languages have complexities not present in logical formalisms that make writing hierarchical grammars difficult. We introduce techniques to handle these complexities, showing how to construct a schema-dependent grammar with minimal over-generation. We analyze these techniques on ATIS and Spider, two challenging text-to-SQL datasets, demonstrating that they yield 14--18\% relative reductions in error.
Subfields: Artificial Intelligence, Information Systems and Management
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mohit Bansal
- Caiming Xiong
- Jonathan Berant
- Luke S. Zettlemoyer
- Richard Socher
- Graham Neubig
- Percy Liang
- Hannaneh Hajishirzi
In Information Systems and Management
- Torsten Scholak
- Jonathan Berant
- Richard Socher
- Caiming Xiong
- Dzmitry Bahdanau
- Matt Gardner
- Nathan Schucher
Citations in the corpus
- Total Recall: a Customized Continual Learning Method for Neural Semantic Parsers (2021, background): “, 2019) or sequences of parse actions (Chen et al., 2018; Cheng et al., 2019; Lin et al., 2019; Zhang et al., 2019; Yin and Neubig, 2018; Chen et al., 2018; Guo et al., 2019; Wang et al., 2020a; Li et al., 2021).”
- PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models (2021, background): “For a while now it has been possible to restrict auto-regressive decoding to only those token sequences that correctly parse to SQL abstract syntax trees (Yin and Neubig, 2018; Lin et al., 2019; Wang et al., 2020).”
- Global Reasoning over Database Structures for Text-to-SQL Parsing (2019, methodology): “Base Model The base parser is a standard topdown semantic parser with grammar-based decoding (Xiao et al., 2016; Yin and Neubig, 2017; Krishnamurthy et al., 2017; Rabinovich et al., 2017; Lin et al., 2019).”
- Photon: A Robust Cross-Domain Text-to-SQL System (2020, background): “Recent advances in neural natural language processing (Sutskever et al., 2014; Dong and Lapata, 2016; See et al., 2017a; Liang et al., 2017; Lin et al., 2019; Bogin et al., 2019a), pre-training (De-vlin et al., 2019; Hwang et al., 2019), and the availability of large-scale supervised datasets…”
- Few-Shot Semantic Parsing for New Predicates (2021, methodology): “At the core of most recent models (Chen et al., 2018; Cheng et al., 2019; Lin et al., 2019; Zhang et al., 2019b; Yin and Neu-big, 2018) is S EQ 2S EQ with attention (Bahdanau et al., 2014) by formulating the task as a machine translation problem.”