DTS-SQL: Decomposed Text-to-SQL with Small Large Language Models

Leading models for the text-to-SQL task heavily rely on proprietary Large Language Models (LLMs), posing concerns over data privacy.Closing the performance gap between small open-source models and large proprietary models is crucial to mitigate this reliance.To this end, we introduce a novel two-stage fine-tuning approach that decomposes the task into two simpler tasks.Through comprehensive evaluation on three large cross-domain datasets and two small LLMs, we show that this approach improves execution accuracy by 3 to 7 percent, effectively aligning the performance of open-source models with their proprietary counterparts.Our proposed method has achieved 60.31% execution accuracy on BIRD hold-out test set, which is the highest performance among methods using 7B parameter models.