让大模型按复杂度自动选择推理路径，提升文本转SQL准确率
DecoSearch: Complexity-Aware Routing and Plan-Level Repair for Text-to-SQL
- 根据问题复杂度动态分配直接生成或分步分解策略
- 在BIRD数据集上达70.53%执行准确率，比无训练基线更高
- 无需修改模型即可提升各类细调模型性能，适合实际部署
大型语言模型在自然语言转SQL方面表现优异，但在需要多步、数据感知推理的复杂查询上仍存在缺陷。本文提出DecoSearch，一种无需训练的框架，通过路由机制将每个查询分配到合适的推理层级。轻量级模式选择器先筛选出相关表和列；语言模型判断器决定是否需分解：简单问题走直接生成路径，复杂问题则升级为由原子子问题组成的有向无环图（DAG），每一步由特定的SQL生成模块解决。检索增强生成（RAG）组件用语义相似的训练样例支撑分解过程，拓扑重构器在执行失败时修正推理计划，区分错误来源。DecoSearch在BIRD上实现70.53%的执行准确率，在Spider上达到88.31%，使用DeepSeek骨干模型，优于所有无训练基线，且消耗的令牌数少一个数量级。该框架具有模型无关性，无需修改即可一致提升微调的SQL生成模型性能。
原文摘要 · Abstract (English)
Large Language Models (LLMs) have demonstrated remarkable capabilities in translating natural language to SQL, yet existing methods still falter on complex queries requiring multi-step, data-aware reasoning. We introduce DecoSearch, a training-free framework that addresses this by routing each query to the appropriate level of reasoning effort. A lightweight Schema Selector first prunes the full database schema to the relevant tables and columns. An LLM Judger then decides whether the question requires decomposition: straightforward questions follow a direct generation path and complex ones are escalated to a Directed Acyclic Graph (DAG) of atomic sub-questions, each solved by a targeted SQL generation step. A RAG component grounds the decomposer with semantically similar training examples, and a Topology Refiner restructures the reasoning plan when execution failures signal a flawed decomposition rather than a fixable SQL error. DecoSearch achieves 70.53% execution accuracy on BIRD and 88.31% on Spider with a DeepSeek backbone, surpassing all training-free baselines while consuming an order of magnitude fewer tokens than competing methods. It also functions as a model-agnostic wrapper, consistently improving fine-tuned SQL generation backbones without any modification to the pipeline.
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。