GitHub

QDA-SQL: Questions Enhanced Dialogue Augmentation for Multi-Turn Text-to-SQL

QDA-SQL leverages large language models (LLMs) to generate multi-turn dialogue samples with diverse question types for Text-to-SQL data augmentation. Given a set of question examples and databases, it ...
IEEE

SAGE: Synthetic-Augmented Generation with Error Feedback for Text-to-SQL

Abstract: The development of text-to-SQL systems has been constrained by the scarcity of high-quality training data. Due to the high cost and limited scalability of manual annotation, researchers have ...
GitHub

SQLDRILLER: Automated Validating and Fixing of Text-to-SQL Translation with Execution Consistency

SQLDRILLER is a Text-to-SQL translation correctness checking tool. It introduces and applies Execution Consistency to automatically discover and fix errors in Text-to-SQL datasets, and in Text-to-SQL ...
IEEE

Text to SQL Transformation Using LLM: A Comparative Research of T5, Seq2Seq, and SQLNet Models

Abstract: The transformation of natural language text into SQL queries is a critical task in the domain of natural language processing and database management. This paper presents a comparative ...