Abstract Current NL2SQL systems degrade sharply when confronted with practical constraints such as limited prompt length and the inability to fine-tune large language models (LLMs). Performance drop is especially pronounced in complex databases, where inaccurate schema linking, vague value conditions, and weak self-correction dominate the error surface. We propose Dynamic-SQL, an adaptive framework that couples multi-path chain-of-thought fusion with execution-based feedback correction. A dense–sparse hybrid vector space is first constructed to dynamically retrieve relevant schema elements, and an LLM is leveraged to generate an explicit schema subgraph. Real-value and few-shot exemplars are then injected to enrich the prompt and sharpen value conditioning. Multiple candidate SQL statements are produced via diverse reasoning paths; their chains of thought are fused to cover latent semantic interpretations, and execution feedback is exploited for iterative self-correction until convergence. On the BIRD benchmark, Dynamic-SQL, powered by the open source qwen2.5-coder-32b-instruct, reduces the average prompt length by 50.83% , raises strict schema-linking recall from 72.63% to 90.66% , and achieves 63.23% execution accuracy. By systematically addressing schema linking, exemplar augmentation, multi-path fusion reasoning, and self-correction, the framework offers a transferable paradigm for deploying LLMs in complex database querying scenarios.
Building similarity graph...
Analyzing shared references across papers
Loading...
Hongbin Hao
Xin Zheng
Xuhong Yu
Scientific Reports
Building similarity graph...
Analyzing shared references across papers
Loading...
Hao et al. (Wed,) studied this question.
www.synapsesocial.com/papers/69e1cf985cdc762e9d8588af — DOI: https://doi.org/10.1038/s41598-026-47693-2