@inproceedings{luo-etal-2026-qbridge,
    title = "{QB}ridge: Bridging Natural Language and {SQL} via Gold Query Rewriting with Agentic Refinement",
    author = "Luo, Zhensheng  and
      Wu, Sai  and
      Qiu, Yuan  and
      Yao, Chang  and
      Chen, Gang  and
      Tang, Xiu",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.402/",
    pages = "8915--8933",
    ISBN = "979-8-89176-390-6",
    abstract = "Natural language to SQL (NL2SQL) provides an intuitive interface for querying structured data, yet real user questions are often noisy, ambiguous, and weakly grounded to database semantics.As a result, token-level schema linking and single-pass SQL decoding can be brittle: small misunderstandings in language or schema grounding may propagate into incorrect generation.We present \textit{QBridge}, an agentic, feedback-driven NL2SQL framework based on a \textit{Refined Gold Query Paradigm}, which bridges natural language and SQL via \textit{Gold Query}{---}a structured, SQL-aligned intermediate representation.A core insight of \textit{QBridge} is \textit{Distilled Back-Translation (DBT)} for SL-independent rewriting.DBT converts SQL-grounded supervision into \textit{execution-verified} Gold-Query-style rewrites from a teacher model, and distills a lightweight, plug-and-play rewriter that generates schema-aware rewrites without requiring explicit schema linking at inference.\textit{QBridge} then (i) verifies and conservatively refines the rewrite into a high-fidelity \textit{Refined Gold Query}, and (ii) refines the generated SQL with dual feedback from execution validity and semantic consistency, enabling interpretable self-correction while remaining compatible with diverse SQL backbones.Extensive experiments on Spider, BIRD, and three robustness variants demonstrate that \textit{QBridge} consistently improves zero-shot NL2SQL, outperforming strong prompting and agentic baselines while showing strong robustness and generalization. Code and data are available at \url{https://github.com/WannaBSteve/QBridge}."
}