Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL
This addresses the bottleneck of producing correct SQL for complex queries in text-to-SQL systems, with incremental improvements in efficiency and scalability.
The researchers tackled the challenge of generating accurate, executable SQL from natural language by developing Arctic-Text2SQL-R1, a reinforcement learning framework that uses a simple execution-based reward signal, achieving state-of-the-art execution accuracy across six benchmarks and outperforming prior 70B-class systems with a 7B model.
Translating natural language into SQL (Test2SQL) is a longstanding challenge at the intersection of natural language understanding and structured data access. While large language models (LLMs) have significantly improved fluency in SQL generation, producing correct and executable SQL--particularly for complex queries--remains a bottleneck. We present Arctic-Text2SQL-R1, a reinforcement learning (RL) framework and model family designed to generate accurate, executable SQL using a lightweight reward signal based solely on execution correctness. Our approach avoids brittle intermediate supervision and complex reward shaping, promoting stable training and alignment with the end task. Combined with carefully curated data, strong supervised initialization, and effective training practices, Arctic-Text2SQL-R1 achieves state-of-the-art execution accuracy across six diverse Test2SQL benchmarks, including the top position on the BIRD leaderboard. Notably, our 7B model outperforms prior 70B-class systems, highlighting the framework's scalability and efficiency. We further demonstrate inference-time robustness through simple extensions like value retrieval and majority voting. Extensive experiments and ablation studies offer both positive and negative insights, providing practical guidance for future Test2SQL research.