RingSQL生成高质量文本到SQL数据，解决强化学习训练中数据稀缺问题。
RingSQL: Schema-Independent Synthetic Data Generation for Text-to-SQL Reinforcement Learning
- 用通用查询模板结合LLM改写，实现无需依赖数据库结构的合成数据生成
- 在多个模型和基准上提升强化学习性能，平均准确率达69.8%
- 适合需要可靠合成数据的文本到SQL研究者与开发者
近期文本到SQL的研究进展得益于更大模型、更优数据集及新训练方法（如RLVR）。但高质量训练数据仍稀缺，而RLVR对噪声数据尤其敏感，错误数据会带来虚假奖励。人工构建成本高，现有合成方法在可靠性与可扩展性间权衡：基于模板的方法虽保证SQL正确性，但需针对特定模式设计模板且多样性不足；基于大模型的方法可扩展性强，却缺乏质量保障。本文提出RingSQL，一种融合无模式查询模板与大模型自然语言改写的混合框架。通过以完整模板问题为基础生成问题，确保在各种查询复杂度下保持问题-查询一致性，这是纯大模型方法无法实现的。RingSQL生成的数据集是唯一能在所有测试模型架构与基准上提升RLVR训练表现的合成数据集，平均准确率达69.8%，优于次佳合成数据集2.1个百分点，并超越Spider和BIRD中的真人标注数据。代码与数据已开源。
原文摘要 · Abstract (English)
Recent advances in text-to-SQL have been driven by larger models, better datasets, and new training methods like RLVR. However, progress remains limited by scarce high-quality training data, a problem RLVR is especially sensitive to since noisy data can produce spurious rewards. Manual data creation is expensive, and existing synthetic methods trade off reliability for scalability: template-based approaches guarantee correct SQL but need schema-specific templates and lack diversity, while LLM-based generation scales easily but lacks quality guarantees. We introduce RingSQL, a hybrid framework for generating question-SQL pairs that combines schema-independent query templates with LLM-based paraphrasing of natural language questions. By grounding question generation in complete template questions, RingSQL preserves question-query correctness across all levels of query complexity, a property purely LLM-based methods fail to maintain. RingSQL also produces the only synthetic dataset that improves RLVR training performance across all tested model architectures and benchmarks, achieving 69.8% average accuracy and surpassing both the next-best synthetic dataset by 2.1% and human-annotated data from Spider and BIRD. Code and data are available at https://github.com/nu-c3lab/RingSQL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。