PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 6, 20260 citationsOpen Access

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

View Full Paper
YWYao WangHLHuiyong LvYQYurong Qian

Key Points

  • The main aim is to enhance the accuracy of SQL queries generated from natural language inputs in Chinese.
  • Developed a dual-model framework separating intent recognition from SQL generation.
  • Implemented a seven-category intent classification system.
  • Utilized large language models for intent recognition and information extraction.
  • Employed a three-level hierarchical backtracking strategy to improve robustness.
  • Reduced semantic-program interference leading to lower structural errors in SQL.
  • Improved intent classification accuracy and query generation performance.
  • Enhanced robustness in handling different types of errors during SQL generation.

Abstract

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 (simpleₛelect, countquery, filterquery, maxₘinquery, sortquery, joinquery, groupbyquery) 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.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/69aa7037531e4c4a9ff59c80https://doi.org/10.3390/ai7030091
Ask AI
Helpful
Bookmark
Share
View Full Paper