Semantic Enhanced Text-to-SQL Parsing via Iteratively Learning Schema Linking Graph

TL;DR

This paper introduces ISESL-SQL, a framework that enhances Text-to-SQL parsing by iteratively learning a semantic schema-linking graph, improving generalization to new databases and handling synonym variations.

Contribution

The paper proposes a novel iterative schema linking approach using PLMs and deep graph learning, with an auxiliary graph regularization task for better schema understanding.

Findings

Consistently outperforms baseline methods on three benchmarks.

Demonstrates strong generalizability to new databases.

Shows robustness in handling lexical variations like synonyms.

Abstract

The generalizability to new databases is of vital importance to Text-to-SQL systems which aim to parse human utterances into SQL statements. Existing works achieve this goal by leveraging the exact matching method to identify the lexical matching between the question words and the schema items. However, these methods fail in other challenging scenarios, such as the synonym substitution in which the surface form differs between the corresponding question words and schema items. In this paper, we propose a framework named ISESL-SQL to iteratively build a semantic enhanced schema-linking graph between question tokens and database schemas. First, we extract a schema linking graph from PLMs through a probing procedure in an unsupervised manner. Then the schema linking graph is further optimized during the training process through a deep graph learning method. Meanwhile, we also design an auxiliary task called graph regularization to improve the schema information mentioned in the schema-linking graph. Extensive experiments on three benchmarks demonstrate that ISESL-SQL could consistently outperform the baselines and further investigations show its generalizability and robustness.