RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQL

By Haoyang Li, Jing Zhang, Cuiping Li, Hong Chen. AAAI Conference on Artificial Intelligence, 2023

379 citations recorded by the source corpus.

One of the recent best attempts at Text-to-SQL is the pre-trained language model. Due to the structural property of the SQL queries, the seq2seq model takes the responsibility of parsing both the schema items (i.e., tables and columns) and the skeleton (i.e., SQL keywords). Such coupled targets increase the difficulty of parsing the correct SQL queries especially when they involve many schema items and logic operators. This paper proposes a ranking-enhanced encoding and skeleton-aware decoding framework to decouple the schema linking and the skeleton parsing. Specifically, for a seq2seq encoder-decode model, its encoder is injected by the most relevant schema items instead of the whole unordered ones, which could alleviate the schema linking effort during SQL parsing, and its decoder first generates the skeleton and then the actual SQL query, which could implicitly constrain the SQL parsing. We evaluate our proposed framework on Spider and its three robustness variants: Spider-DK, Spider-Syn, and Spider-Realistic. The experimental results show that our framework delivers promising performance and robustness. Our code is available at https://github.com/RUCKBReasoning/RESDSQL.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/6d51f4b220cb2c8321dc5f9755b7d66f10f1cad6

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

1. Jingren Zhou
2. Bolin Ding
3. Xiuyu Sun
4. Yaliang Li

Citations in the corpus

• Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation (2023, methodology): “Most previous works [14, 19, 20, 42, 50] focus on extracting the question-to-SQL patterns and generalizing them by training an encoder-decoder model with Text-to-SQL corpus.”
• Retrieval-Augmented Generation for AI-Generated Content: A Survey (2024, neutral): “RES-DSQL [282] ranks schemas using cross-encoder, then includes ranked schemas into prompts to generate SQL skeleton and SQL query.”
• CodeS: Towards Building Open-source Language Models for Text-to-SQL (2024, neutral): “The training and evaluation processes for the schema item classifier are conducted in accordance with [36].”
• Finding the Right Tables and Columns: A Benchmark and Corpus-Adaptive Embeddings for SQL Schema Retrieval (2026, unknown): “Selecting the relevant schema elements first, a form of schema linking (Li et al., 2023a), is therefore a pre-requisite for database tasks over large schemas, and it constitutes a natural retrieval problem: ranking schema documents by their relevance to a query.”