On the Potential of Lexico-logical Alignments for Semantic Parsing to SQL Queries
By Tianze Shi, Chen Zhao, Jordan L. Boyd-Graber, Hal Daumé Iii, Lillian Lee. 2020
90 citations recorded by the source corpus.
Large-scale semantic parsing datasets annotated with logical forms have enabled major advances in supervised approaches. But can richer supervision help even more? To explore the utility of fine-grained, lexical-level supervision, we introduce SQUALL, a dataset that enriches 11,276 WIKITABLEQUESTIONS English-language questions with manually created SQL equivalents plus alignments between SQL and question fragments. Our annotation enables new training possibilities for encoderdecoder models, including approaches from machine translation previously precluded by the absence of alignments. We propose and test two methods: (1) supervised attention; (2) adopting an auxiliary objective of disambiguating references in the input queries to table columns. In 5-fold cross validation, these strategies improve over strong baselines by 4.4% execution accuracy. Oracle experiments suggest that annotated alignments can support further accuracy gains of up to 23.9%.
Subfields: Artificial Intelligence, Natural Language Processing
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
- Luke S. Zettlemoyer
- Wen-tau Yih
- Caiming Xiong
- Percy Liang
- Mohit Bansal
- Dragomir R. Radev
- Jianfeng Gao
- Graham Neubig
In Natural Language Processing
- Caiming Xiong
- Luke S. Zettlemoyer
- Wen-tau Yih
- Dragomir R. Radev
- Richard Socher
- Percy Liang
- Mirella Lapata
- Jianfeng Gao
Citations in the corpus
- Reasoning Like Program Executors (2022, methodology): “Concretely, by instantiating SQL templates from S QUALL (Shi et al., 2020)”
- Binding Language Models in Symbolic Languages (2022, methodology): “It gains advantages in interpretability and robustness compared with End-to-End and Chain-of-Thought methods, but still suffers from the fixed grammar of the pre-defined programming language, making it inherently limited in coverage (Shi et al., 2020b).”
- LEVER: Learning to Verify Language-to-Code Generation with Execution (2023, methodology): “The ability of mapping natural language to executable code is the cornerstone of a variety AI applications such as database interfaces (Pasupat & Liang, 2015; Yu et al., 2018; Shi et al., 2020), robotics control (Zhou et al.”