Memo-SQL: Structured Decomposition and Experience-Driven Self-Correction for Training-Free NL2SQL
This work addresses efficiency and robustness issues in natural language to SQL conversion for database users, representing a strong specific gain rather than a foundational advancement.
The paper tackled the limitations of existing NL2SQL systems, such as reliance on correct examples only and inefficient decomposition, by introducing Memo-SQL, a training-free framework that uses structured decomposition and experience-driven self-correction, achieving 68.5% execution accuracy on BIRD with over 10 times fewer resources.
Existing NL2SQL systems face two critical limitations: (1) they rely on in-context learning with only correct examples, overlooking the rich signal in historical error-fix pairs that could guide more robust self-correction; and (2) test-time scaling approaches often decompose questions arbitrarily, producing near-identical SQL candidates across runs and diminishing ensemble gains. Moreover, these methods suffer from a stark accuracy-efficiency trade-off: high performance demands excessive computation, while fast variants compromise quality. We present Memo-SQL, a training-free framework that addresses these issues through two simple ideas: structured decomposition and experience-aware self-correction. Instead of leaving decomposition to chance, we apply three clear strategies, entity-wise, hierarchical, and atomic sequential, to encourage diverse reasoning. For correction, we build a dynamic memory of both successful queries and historical error-fix pairs, and use retrieval-augmented prompting to bring relevant examples into context at inference time, no fine-tuning or external APIs required. On BIRD, Memo-SQL achieves 68.5% execution accuracy, setting a new state of the art among open, zero-fine-tuning methods, while using over 10 times fewer resources than prior TTS approaches.