LIMIT：少即是多——面向Text-to-SQL的指令微调
LIMIT: Less Is More for Instruction Tuning in Text-to-SQL
- Zhejiang University（浙江大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
LIMIT提出以数据为中心的框架，通过难度过滤、思维链合成、质量评分和遗传算法，仅用796和863个样本实现100%表覆盖，使Qwen3-8B在BIRD和Spider上达到69.1%和88.9%执行准确率，超越20倍数据训练的方法，证明数据质量比规模更重要。
AI中文摘要：
大语言模型通过增强推理的微调在Text-to-SQL任务上取得了显著进展，然而现有方法主要依赖大规模指令语料库，并假设数据规模驱动性能提升。我们通过探究一个基本问题来挑战这一范式：有效的Text-to-SQL指令微调所需的最小数据量是多少？我们提出LIMIT（少即是多——面向Text-to-SQL的指令微调），一个以数据为中心的框架，证明了当示例被策略性地选择时，强大的数据库推理能力可以从极其紧凑的训练集中涌现。LIMIT通过四个阶段运作：难度感知过滤，识别模型学习前沿内的样本；思维链合成与基于一致性的选择；通过LLM-as-judge进行多维质量评分；以及联合最大化模式覆盖率和样本质量的遗传算法优化。在BIRD和Spider基准上，LIMIT仅选择796和863个样本，同时实现100%的表覆盖率，使Qwen3-8B达到69.1%和88.9%的执行准确率。这一结果超越了使用20倍数据训练的方法，并在开源方法中确立了新的最先进水平。我们的发现表明，精心的数据策展而非规模，是高效Text-to-SQL学习的关键。
英文摘要：
Large language models have achieved remarkable progress on Text-to-SQL through reasoning-enhanced fine-tuning, yet existing approaches predominantly rely on massive instruction corpora under the assumption that scale drives performance. We challenge this paradigm by investigating a fundamental question: what is the minimal data requirement for effective Text-to-SQL instruction tuning? We propose LIMIT(Less Is More for Instruction Tuning in Text-to-SQL), a data-centric framework that demonstrates strong database reasoning can emerge from an extremely compact training set when examples are strategically selected. LIMIT operates through four stages: difficulty-aware filtering that identifies samples within the model's learning frontier, chain-of-thought synthesis with consistency-based selection, multi-dimensional quality scoring via LLM-as-judge, and genetic algorithm optimization that jointly maximizes schema coverage and sample quality. On the BIRD and Spider benchmark, LIMIT selects only 796 and 863 samples while achieving 100% table coverage, enabling Qwen3-8B to reach 69.1% and 88.9% execution accuracy.This result surpasses methods trained on 20 times more data and establishes a new state-of-the-art among open-source approaches. Our findings suggest that careful data curation, rather than scale, is the key to efficient Text-to-SQL learning.
