HES-SQL: Hybrid Reasoning for Efficient Text-to-SQL with Structural Skeleton Guidance
This work addresses the need for robust natural language interfaces to databases by balancing semantic accuracy with computational efficiency, representing a new paradigm for Text-to-SQL systems.
The paper tackled the problem of generating efficient and accurate SQL queries from natural language by introducing HES-SQL, a hybrid training framework that integrates thinking-mode-fused supervised fine-tuning with Group Relative Policy Optimization, achieving execution accuracies of 79.14% on BIRD and 54.9% on KaggleDBQA benchmarks with efficiency gains of 11% to 20% relative to baselines.
We present HES-SQL, a novel hybrid training framework that advances Text-to-SQL generation through the integration of thinking-mode-fused supervised fine-tuning (SFT) with Group Relative Policy Optimization (GRPO). Our approach introduces three key innovations: (1) a skeleton-completeness scoring mechanism that enhances preference alignment between generated queries and optimal SQL structures; (2) a query-latency-aware reward system that incentivizes the generation of computationally efficient SQL queries; (3) a self-distillation process for thinking-mode completion that prevents degradation of the model's reasoning capabilities. This framework enables hybrid thinking models to switch between reasoning and non-reasoning modes while improving SQL query accuracy and execution efficiency. Experimental evaluation, conducted on MySQL 8.0 and SQLite 3.42 under controlled single-user conditions, demonstrates that HES-SQL achieves competitive performance with execution accuracies of 79.14\% and 54.9\% on the BIRD and KaggleDBQA benchmarks, respectively. Query latency is measured as the end-to-end execution time of generated queries on the DBMS, averaged over multiple runs to mitigate variance. Efficiency gains range from 11\% to 20\% relative to supervised baselines. Our results establish a new paradigm for Text-to-SQL systems that effectively balances semantic accuracy with computational efficiency through execution-informed reinforcement learning (RL). The proposed methodology has significant implications for developing robust natural language interfaces to databases and can be extended to broader structured generation tasks requiring both correctness and efficiency optimization.