cs.AIApr 23, 2026

SemanticAgent: A Semantics-Aware Framework for Text-to-SQL Data Synthesis

Authors: Qiang GaoZhenping LiAnqi ZhuoYingxiao ZhaoWeibo GengXiaosong Li

Organizations: Center of Information Research, Academy of Military Science, Beijing 100142, China · School of Mathematics and Physics, University of Science and Technology Beijing, Beijing 100083, China

Abstract

Existing text-to-SQL synthesis pipelines still conflate executability with semantic validity: syntactic checks and execution-based validation can retain queries that execute successfully while violating database semantics. To address these limitations, we propose SemanticAgent, a semantic-aware synthesis framework. SemanticAgent organizes synthesis around three specialized modules: an analyzer, a synthesizer, and a verifier. Through a three-stage protocol of semantic analysis, stepwise synthesis, and diagnostic refinement, SemanticAgent transforms execution-based validation alone into a traceable reasoning process. Our framework generates synthetic data that consistently outperforms prior synthesis methods under semantic-quality evaluation, leading to stronger downstream fine-tuning performance, especially on semantically demanding benchmarks.

Explore similar work

CardsList
  1. PExA: Parallel Exploration Agent for Complex Text-to-SQL

    Apr 24, 2026Tanmay Parekh, Ella Hofmann-Coyle, Shuyi Wang +3Text-To-SqlSql