SemanticAgent: A Semantics-Aware Framework for Text-to-SQL Data Synthesis
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911618312437760 |
|---|---|
| author | Gao, Qiang Li, Zhenping Zhuo, Anqi Zhao, Yingxiao Geng, Weibo Li, Xiaosong |
| author_facet | Gao, Qiang Li, Zhenping Zhuo, Anqi Zhao, Yingxiao Geng, Weibo Li, Xiaosong |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_21414 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SemanticAgent: A Semantics-Aware Framework for Text-to-SQL Data Synthesis Gao, Qiang Li, Zhenping Zhuo, Anqi Zhao, Yingxiao Geng, Weibo Li, Xiaosong Artificial Intelligence I.2.7; I.2.4 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. |
| title | SemanticAgent: A Semantics-Aware Framework for Text-to-SQL Data Synthesis |
| topic | Artificial Intelligence I.2.7; I.2.4 |
| url | https://arxiv.org/abs/2604.21414 |