EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CIR-SQL: A Dual-Model Intent Recognition Framework for Chinese Text-to-SQL

Yao Wang, Huiyong Lv, Yurong Qian

AI · 04.03.2026 · 7 (3) · 91

DOI: 10.3390/ai7030091 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AI, online erschienen am 04.03.2026. Verfasst von Yao Wang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Key Research and Development Project in Xinjiang Uygur Autonomous Region.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In Industry 4.0 environments, operators and production managers frequently query industrial databases for production monitoring, quality control, and equipment maintenance using natural language. Existing Chinese NL2SQL systems often process semantic, program, and schema information in a single encoder, which leads to semantic-program interference and frequent structural or schema errors in the generated SQL. We present CIR-SQL, a dual-model framework that separates intent recognition from SQL generation via structured intermediate representations, decoupling semantic understanding from program synthesis. CIR-SQL employs a seven-category intent classification system (simple_select, count_query, filter_query, max_min_query, sort_query, join_query, group_by_query) and leverages large language models for intent recognition and structured information extraction. A three-level hierarchical backtracking strategy (SQL, context, intent) further improves robustness by correcting different error types. The architecture is particularly suited to Industry 4.0 scenarios where Chinese-speaking operators interact with complex industrial databases containing production data, quality metrics, and equipment status information.

Förderung

  • the Key Research and Development Project in Xinjiang Uygur Autonomous Region

Zitieren

APA

Wang, Y., Lv, H., & Qian, Y. (2026). CIR-SQL: A Dual-Model Intent Recognition Framework for Chinese Text-to-SQL. AI, 7(3), 91. https://doi.org/10.3390/ai7030091

Vancouver

Wang Y, Lv H, Qian Y. CIR-SQL: A Dual-Model Intent Recognition Framework for Chinese Text-to-SQL. AI. 2026;7(3):91. doi:10.3390/ai7030091

BibTeX
@article{Wang2026,
  author = {Wang, Yao and Lv, Huiyong and Qian, Yurong},
  title = {CIR-SQL: A Dual-Model Intent Recognition Framework for Chinese Text-to-SQL},
  journal = {AI},
  year = {2026},
  volume = {7},
  number = {3},
  pages = {91},
  doi = {10.3390/ai7030091}
}
Kostenlos starten →