A Tale of Two Linkings: Dynamically Gating between Schema Linking and Structural Linking for Text-to-SQL Parsing

Proceedings of the 28th International Conference on Computational Linguistics, pages 2900–2912 Barcelona, Spain (Online), December 8-13, 2020 2900 A Tale of Two Linkings: Dynamically Gating between Schema Linking and Structural Linking for Text-to-SQL Parsing Sanxing Chen♦ Aidan San♦ Xiaodong Liu♣ Yangfeng Ji♦ ♦University of Virginia ♣Microsoft Research {sc3hn, aws9xm, yangfeng}@virginia.edu xiaodl@microsoft.com Abstract In Text-to-SQL semantic parsing, selecting the correct entities (tables and columns) for the generated SQL query is both crucial and challenging; the parser is required to connect the natural language (NL) question and the SQL query to the structured knowledge in the database. We formulate two linking processes to address this challenge: schema linking which links explicit NL mentions to the database and structural linking which links the entities in the output SQL with their structural relationships in the database schema. Intuitively, the effectiveness of these two linking processes changes based on the entity being generated, thus we propose to dynamically choose between them using a gating mechanism. Integrating the proposed method with two graph neural network-based semantic parsers together with BERT representations demonstrates substantial gains in parsing accuracy on the challenging Spider dataset. Analyses show that our proposed method helps to enhance the structure of the model output when generating complicated SQL queries and offers more explainable predictions. 1 Introduction Semantic parsing, which aims at mapping natural language (NL) utterances to computer understandable logic forms or programming languages, has been an active research topic in the field of natural language processing (NLP) for decades (Zettlemoyer and Collins, 2005; Liang et al., 2011). Although a variety of logic forms have been studied by researchers, Text-to-SQL has particularly attracted a large amount of attention due to the desire of natural language interfaces to database (NLIDB) (Warren and Pereira, 1982; Zelle and Mooney, 1996; Dong, 2019) for both scientific and industrial reasons. Recently, there is a growing interest in neural based Text-to-SQL semantic parsing, thanks to the development of new evaluation paradigms and datasets (Iyer et al., 2017; Zhong et al., 2017; Yu et al., 2018; Yu et al., 2019). Text-to-SQL parsing requires strict structured prediction due to its application scenario where the output SQL will be sent to an executor program directly. To enhance the capacity of an auto-regressive model to capture structural information, current state-of-the-art semantic parsers usually adopt a grammar-based decoder (Xiao et al., 2016; Yin and Neubig, 2017; Krishnamurthy et al., 2017). Rather than directly generating the tokens in a traditional sequence-to-sequence manner, grammar-based decoders produce a sequence of production rules to construct an abstract syntax tree (AST) of the corresponding SQL. As the grammar constraints narrows down the search space to only grammatically valid ASTs, those parsers can usually generate well-formed SQL skeletons (Guo et al., 2019; Bogin et al., 2019a). However, it is still difficult for current state-of-the-art models to fill in the skeletons with semantically correct entities, especially when they are required to generalize to unseen DB schemas (Yu et al., 2018; Suhr et al., 2020). To predict the correct entity, the model …

Content truncated to 500 words