SQL-to-Text Generation with Weighted-AST Few-Shot Prompting

2025-11-17

AI Summary

SQL-to-text generation aims to enhance non-technical users’ comprehension of database queries, yet existing approaches suffer from insufficient logical fidelity due to semantic drift and lexical diversity. To address this, we propose the AST-Guided Weighted Retrieval (AWR) framework: first, SQL queries are parsed into Abstract Syntax Trees (ASTs) to explicitly encode structural semantics; second, a weighted AST similarity retrieval mechanism is introduced, dynamically learning node importance weights to enable structure-aware few-shot prompting; finally, a large language model generates natural-language descriptions conditioned on retrieved exemplars. Evaluated on Spider, SParC, and CoSQL benchmarks, AWR achieves up to 17.24% absolute improvement in exact-match accuracy. Human evaluation confirms significant gains in semantic consistency and descriptive fluency over baselines, while maintaining efficient inference.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Data Mining & Knowledge Management: Intelligent Query Processing
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

SQL-to-Text generation aims at translating structured SQL queries into natural language descriptions, thereby facilitating comprehension of complex database operations for non-technical users. Although large language models (LLMs) have recently demonstrated promising results, current methods often fail to maintain the exact semantics of SQL queries, particularly when there are multiple possible correct phrasings. To address this problem, our work proposes Weighted-AST retrieval with prompting, an architecture that integrates structural query representations and LLM prompting. This method retrieves semantically relevant examples as few-shot prompts using a similarity metric based on an Abstract Syntax Tree (AST) with learned feature weights. Our structure-aware prompting technique ensures that generated descriptions are both fluent and faithful to the original query logic. Numerous experiments on three benchmark datasets - Spider, SParC, and CoSQL show that our method outperforms the current baselines by up to +17.24% in execution Accuracy (EX), performs superior in Exact Match (EM) and provides more consistent semantic fidelity when evaluated by humans, all while preserving competitive runtime performance. These results demonstrate that Weighted-AST prompting is a scalable and effective method for deriving natural language explanations from structured database queries.

Problem

• Translating SQL queries into natural language for non-technical users
• Maintaining exact SQL semantics across multiple phrasing variations
• Improving semantic fidelity in SQL-to-text generation using structure-aware methods

Innovation

• Weighted-AST retrieval for semantic example selection
• Structure-aware prompting with learned feature weights
• Integrating Abstract Syntax Trees with LLM few-shot learning