重思 Text-to-SQL：大语言模型架构与策略综述
原标题：Reimagining Text-to-SQL: A Survey of Large Language Model Architectures and Strategies
AI 摘要
该论文是一篇关于大语言模型驱动 Text-to-SQL 的系统性综述，覆盖 2022 至 2026 年间的 100 多篇论文与技术资源。作者提出五类范式分类（上下文学习、监督与参数高效微调、智能体与工具增强编排、检索增强生成、强化学习），并按模型粒度、监督程度和推理风格刻画各类方法。文章还分析了评估指标、数据集与开放挑战，包括领域泛化、效率、真实部署及语义层作用，并讨论了敏感数据库访问下的负责任 AI 实践。
正文节选
Reimagining Text-to-SQL: A Survey of Large Language Model Architectures and Strategies Abstract Text-to-SQL systems translate natural language questions into executable SQL queries, democratizing access to relational data. Early approaches relied on handcrafted rules and template matching, while subsequent neural methods introduced encoder–decoder architectures, schema linking, and constrained decoding to handle increasingly complex queries. The emergence of large language models (LLMs), exempli