Elamur Research⌕
☰
home / papers / 2026 / Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL

Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL

Paper · 2026 · Zhewei Yao, Guoheng Sun, Łukasz Borchmann et al.

Summary

The paper addresses a persistent weakness of LLM-based Text2SQL: models write fluent SQL, but complex queries with multi-table joins and nested logic often fail to execute or return wrong results. The authors propose Arctic-Text2SQL-R1, a reinforcement learning framework and model family built on vanilla GRPO. Unlike competing systems, which combine many reward components, it uses a deliberately minimal reward. A query earns full credit when its execution output matches the ground truth, a small credit when it merely runs, and nothing otherwise. This RL stage is combined with filtered and synthetic training data, strong supervised starting checkpoints and careful prompting. Across six Text2SQL benchmarks, the models beat both SQL-specialized and general LLMs such as GPT-4o and DeepSeek-V3. The 32B model places among the top entries on the BIRD leaderboard and ranks first among systems with publicly documented details. The work suggests that simple, verifiable rewards can outperform elaborate reward engineering, which tends to invite reward hacking.

Key points

  • Reward scheme: 1 for an execution match with the gold query, 0.1 for executable SQL, and 0 otherwise. Training uses vanilla GRPO, chosen to limit confounding factors.
  • On BIRD-test, the 32B model reaches 71.83% execution accuracy, rising to 73.84% with self-consistency. The 7B model matches or surpasses ExCoT-70B and GPT-4o.
  • In a controlled comparison from Qwen2.5-Coder-14B-Instruct, it outperforms Reasoning-SQL on BIRD-dev (+2.28), Spider-test (+5.77) and Spider-DK (+2.07). It also beats SQL-R1 and Think2SQL with 7B bases, including OmniSQL-7B.
  • Inference-time value retrieval and majority voting (8 samples) together raise the 32B model on BIRD-dev from 70.5 to 71.5.
  • Released resources include the Snowflake/Arctic-Text2SQL-R1-7B model on Hugging Face and the ArcticTraining code. A noted limitation is that execution-only rewards can still credit semantically wrong queries that happen to return the correct result.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓University of Maryland, College ParkCollege Park · United States132 works · 25,380 citations🎓University of California San DiegoSan Diego · United States73 works · 51,948 citations🏢Snowflake Inc.Bozeman · United States3 works · 17 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Networks and Communications · 64 works

🧭 Topics 3Research subjects and languages the work is about.Reasoning · 1,426 worksSemantic parsing & meaning representation · 185 worksCode & programming languages · 129 works

Cite

Cite as Zhewei Yao, Guoheng Sun, Łukasz Borchmann, Zheyu Shen, Minghang Deng, Bohan Zhai, Hao Zhang, Ang Li and Yuxiong He. 2026. Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.1345

@inproceedings{yao2026arctic,
  title = {{Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL}},
  author = {Zhewei Yao, Guoheng Sun, Łukasz Borchmann, Zheyu Shen, Minghang Deng, Bohan Zhai, Hao Zhang, Ang Li and Yuxiong He},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1345},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1345}
}

References

62 references, 25 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. I. Androutsopoulos, G. D. Ritchie, and P. Thanisch. 1995. Natural language interfaces to databases — an introduction. Preprint, arXiv:cmp-lg/9503016. find
  2. Quy-Anh Dang and Chris Ngo. 2025. Reinforcement Learning for reasoning in small LLMs: What works and what doesn’t. ArXiv, abs/2503.16219. find
  3. Minghang Deng, Ashwin Ramachandran, Canwen Xu, Lanxiang Hu, Zhewei Yao, Anupam Datta, and Hao Zhang. 2025. ReFoRCE: A text-to-SQL agent with self-refinement, format restriction, and column exploration. Preprint, arXiv:2502.00675. — [arXiv:2502.00675](https://arxiv.org/abs/2502.00675) find
  4. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, and 1 others. 2024. The llama 3 herd of models. arXiv e-prints, pages arXiv–2407.
  5. Ben Eyal, Moran Mahabi, Ophir Haroche, Amir Bachar, and Michael Elhadad. 2023. Semantic decomposition of question and SQL for text-to-SQL parsing. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 13629–13645, Singapore. Association for Computational Linguistics. find
  6. Yujian Gan, Xinyun Chen, and Matthew Purver. 2021. Exploring underexplored limitations of cross-domain text-to-SQL generalization. Preprint, arXiv:2109.05157.
  7. Chang Gao, Chujie Zheng, Xiong-Hui Chen, Kai Dang, Shixuan Liu, Bowen Yu, An Yang, Shuai Bai, Jingren Zhou, and Junyang Lin. 2025. Soft adaptive policy optimization. arXiv preprint arXiv:2511.20347. — [arXiv:2511.20347](https://arxiv.org/abs/2511.20347) find
  8. Dawei Gao, Haibin Wang, Yaliang Li, Xiuyu Sun, Yichen Qian, Bolin Ding, and Jingren Zhou. 2023. Text-to-SQL empowered by Large Language Models: A benchmark evaluation. Preprint, arXiv:2308.15363.
  9. Yingqi Gao, Yifu Liu, Xiaoxia Li, Xiaorong Shi, Yin Zhu, Yiming Wang, Shiqi Li, Wei Li, Yuntao Hong, Zhiling Luo, Jinyang Gao, Liyu Mou, and Yu Li. 2024. A preview of XiYan-SQL: A multi-generator ensemble framework for text-to-SQL. arXiv preprint arXiv:2411.08599. — [arXiv:2411.08599](https://arxiv.org/abs/2411.08599) find
  10. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, and 1 others. 2024. The Llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  11. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, and 1 others. 2025. DeepSeek-R1: Incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948.
  12. Daya Guo, Qihao Zhu, Dejian Yang, Zhenda Xie, Kai Dong, Wentao Zhang, Guanting Chen, Xiao Bi, Yu Wu, YK Li, and 1 others. 2024. DeepSeek-Coder: When the large language model meets programming–the rise of code intelligence. arXiv preprint arXiv:2401.14196. find
  13. Jiaqi Guo, Zecheng Zhan, Yan Gao, Yan Xiao, Jian-Guang Lou, Ting Liu, and Dongmei Zhang. 2019. Towards complex text-to-SQL in cross-domain database with intermediate representation. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4524–4535, Florence,… find
  14. Mingqian He, Yongliang Shen, Wenqi Zhang, Qiuying Peng, Jun Wang, and Weiming Lu. 2025. STaR-SQL: Self-taught reasoner for text-to-SQL. Preprint, arXiv:2502.13550.
  15. Zijin Hong, Zheng Yuan, Hao Chen, Qinggang Zhang, Feiran Huang, and Xiao Huang. 2024. Knowledge-to-SQL: Enhancing SQL generation with data expert LLM. Preprint, arXiv:2402.11517. find
  16. Yiqun Hu, Yiyun Zhao, Jiarong Jiang, Wuwei Lan, Henghui Zhu, Anuj Chauhan, Alexander Hanbo Li, Lin Pan, Jun Wang, Chung-Wei Hang, Sheng Zhang, Jiang Guo, Mingwen Dong, Joseph Lilien, Patrick Ng, Zhiguo Wang, Vittorio Castelli, and Bing Xiang. 2023. Importance of synthesizing high-quality data for te… find
  17. Siming Huang, Tianhao Cheng, Jason Klein Liu, Jiaran Hao, Liuyihan Song, Yang Xu, J Yang, JH Liu, Chenchen Zhang, Linzheng Chai, and 1 others. 2024. OpenCoder: The open cookbook for top-tier code large language models. arXiv preprint arXiv:2411.04905.
  18. Binyuan Hui, Jian Yang, Zeyu Cui, Jiaxi Yang, Dayiheng Liu, Lei Zhang, Tianyu Liu, Jiajun Zhang, Bowen Yu, Keming Lu, and 1 others. 2024. Qwen2.5-Coder technical report. arXiv preprint arXiv:2409.12186.
  19. Albert Q Jiang, Alexandre Sablayrolles, Antoine Roux, Arthur Mensch, Blanche Savary, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Emma Bou Hanna, Florian Bressand, and 1 others. 2024. Mixtral of Experts. arXiv preprint arXiv:2401.04088. find
  20. Komal Kumar, Tajamul Ashraf, Omkar Thawakar, Rao Muhammad Anwer, Hisham Cholakkal, Mubarak Shah, Ming-Hsuan Yang, Phillip H. S. Torr, Fahad Shahbaz Khan, and Salman Khan. 2025. LLM post-training: A deep dive into reasoning large language models. Preprint, arXiv:2502.21321. — [arXiv:2502.21321](https://arxiv.org/abs/2502.21321) find
  21. Nathan Lambert, Jacob Daniel Morrison, Valentina Pyatkin, Shengyi Huang, Hamish Ivison, Faeze Brahman, Lester James Validad Miranda, Alisa Liu, Nouha Dziri, Xinxi Lyu, Yuling Gu, Saumya Malik, Victoria Graf, Jena D. Hwang, Jiangjiang Yang, Ronan Le Bras, Oyvind Tafjord, Chris Wilhelm, Luca Soldaini,… — Tulu 3: Pushing Frontiers in Open Language Model Post-Training
  22. Gyubok Lee, Hyeonji Hwang, Seongsu Bae, Yeonsu Kwon, Woncheol Shin, Seongjun Yang, Minjoon Seo, Jong-Yeup Kim, and Edward Choi. 2022. EHRSQL: A practical text-to-SQL benchmark for electronic health records. Advances in Neural Information Processing Systems, 35:15589–15601. find
  23. Fei Li and H. V. Jagadish. 2014. Constructing an interactive natural language interface for relational databases. Proc. VLDB Endow., 8(1):73–84. find
  24. Haoyang Li, Shang Wu, Xiaokang Zhang, Xinmei Huang, Jing Zhang, Fuxin Jiang, Shuai Wang, Tieying Zhang, Jianjun Chen, Rui Shi, Hong Chen, and Cuiping Li. 2025. OmniSQL: Synthesizing high-quality text-to-SQL data at scale. arXiv preprint arXiv:2503.02240. — [arXiv:2503.02240](https://arxiv.org/abs/2503.02240) find
  25. Jinyang Li, Binyuan Hui, Ge Qu, Jiaxi Yang, Binhua Li, Bowen Li, Bailin Wang, Bowen Qin, Ruiying Geng, Nan Huo, and 1 others. 2024. Can LLM already serve as a database interface? a big bench for large-scale database grounded text-to-SQLs. Advances in Neural Information Processing Systems, 36.
  26. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024. DeepSeek-v3 technical report. arXiv preprint arXiv:2412.19437. find
  27. Anton Lozhkov, Raymond Li, Loubna Ben Allal, Federico Cassano, Joel Lamy-Poirier, Nouamane Tazi, Ao Tang, Dmytro Pykhtar, Jiawei Liu, Yuxiang Wei, and 1 others. 2024. StarCoder 2 and The Stack v2: The next generation. arXiv preprint arXiv:2402.19173. find
  28. Peixian Ma, Xialie Zhuang, Chengjin Xu, Xuhui Jiang, Ran Chen, and Jian Guo. 2025. SQL-R1: Training natural language to SQL reasoning model by reinforcement learning. arXiv preprint arXiv:2504.08600. — [arXiv:2504.08600](https://arxiv.org/abs/2504.08600) find
  29. Yev Meyer, Marjan Emadi, Dhruv Nathawani, Lipika Ramaswamy, Kendrick Boyd, Maarten Van Segbroeck, Matthew Grossman, Piotr Mlocek, and Drew Newberry. 2024. Synthetic-Text-To-SQL: A synthetic dataset for training language models to generate sql queries from natural language prompts. find
  30. Mayank Mishra, Matt Stallone, Gaoyuan Zhang, Yikang Shen, Aditya Prasad, Adriana Meza Soria, Michele Merler, Parameswaran Selvam, Saptha Surendran, Shivdeep Singh, and 1 others. 2024. Granite code models: A family of open foundation models for code intelligence. arXiv preprint arXiv:2405.04324. find
  31. Mistral AI. 2024. Codestral: Mistral AI’s first code generation model. https://mistral.ai/news/co destral. Accessed: May 1, 2025. ↗
  32. OpenAI. 2023. GPT-4 technical report. Accessed: May 1, 2025.
  33. OpenAI and 1 others. 2024. OpenAI o1 system card. Preprint, arXiv:2412.16720.
  34. OpenAI and 1 others. 2025. Competitive programming with Large Reasoning Models. Preprint, arXiv:2502.06807. find
  35. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, and 1 others. 2022. Training language models to follow instructions with human feedback. Advances in neural information processing systems, 35:27730–27744.
  36. Simone Papicchio, Simone Rossi, Luca Cagliero, and Paolo Papotti. 2025. Think2SQL: Reinforce LLM reasoning capabilities for Text2SQL. Preprint, arXiv:2504.15077. find
  37. Mohammadreza Pourreza, Shayan Talaei, Ruoxi Sun, Xingchen Wan, Hailong Li, Azalia Mirhoseini, Amin Saberi, Sercan Arik, and 1 others. 2025. Reasoning-SQL: Reinforcement learning with SQL tailored partial rewards for reasoning-enhanced text-to-SQL. arXiv preprint arXiv:2503.23157. — [arXiv:2503.23157](https://arxiv.org/abs/2503.23157) find
  38. John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. 2017. Proximal policy optimization algorithms. Preprint, arXiv:1707.06347. find
  39. Shao, Yuwei Yan, Qinglong Yang, Yiwen Song, Sijian Ren, Xinyuan Hu, Yu Li, Jie Feng, Chen Gao, and Yong Li. 2025. Towards large reasoning models: A survey of reinforced reasoning with Large Language Models. Preprint, arXiv:2501.09686. find
  40. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, Y.K. Li, Y. Wu, and Daya Guo. 2024. DeepSeekMath: Pushing the limits of mathematical reasoning in open language models. Preprint, arXiv:2402.03300.
  41. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025a. Qwen3 technical report. arXiv preprint arXiv:2505.09388.
  42. Aditi Singh, Akash Shetty, Abul Ehtesham, Saket Kumar, and Tala Talaei Khoei. 2025. A survey of large language model-based generative ai for text-to-SQL: Benchmarks, applications, use cases, and challenges. Preprint, arXiv:2412.05208. — [arXiv:2412.05208](https://arxiv.org/abs/2412.05208) find
  43. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, and 1 others. 2024a. Qwen2.5 technical report. arXiv preprint arXiv:2412.15115. find
  44. Joar Skalse, Nikolaus Howe, Dmitrii Krasheninnikov, and David Krueger. 2022. Defining and characterizing reward gaming. Advances in Neural Information Processing Systems, 35:9460–9471. find
  45. Chang-You Tai, Ziru Chen, Tianshu Zhang, Xiang Deng, and Huan Sun. 2023. Exploring chain-of-thought style prompting for text-to-SQL. Preprint, arXiv:2305.14215. find
  46. Shayan Talaei, Mohammadreza Pourreza, Yu-Chen Chang, Azalia Mirhoseini, and Amin Saberi. 2024. CHESS: Contextual harnessing for efficient SQL synthesis. Preprint, arXiv:2405.16755. ArXiv preprint arXiv:2405.16755. — [arXiv:2405.16755](https://arxiv.org/abs/2405.16755) find
  47. Team Cohere. 2025. Command A: An enterprise-ready large language model. Preprint, arXiv:2504.00698. Accessed: 2025-05-04. — [arXiv:2504.00698](https://arxiv.org/abs/2504.00698) find
  48. An Yang, Beichen Zhang, Binyuan Hui, Bofei Gao, Bowen Yu, Chengpeng Li, Dayiheng Liu, Jianhong Tu, Jingren Zhou, Junyang Lin, Keming Lu, Mingfeng Xue, Runji Lin, Tianyu Liu, Xingzhang Ren, and Zhenru Zhang. 2024b. Qwen2.5-Math technical report: Toward mathematical expert model via self-improvement.…
  49. Dayu Yang, Tianyang Liu, Daoan Zhang, Antoine Simoulin, Xiaoyi Liu, Yuwei Cao, Zhaopu Teng, Xin Qian, Grey Yang, Jiebo Luo, and Julian McAuley. 2025b. Code to think, think to code: A survey on code-enhanced reasoning and reasoning-driven code intelligence in LLMs. Preprint, arXiv:2502.19411.
  50. Jiaxi Yang, Binyuan Hui, Min Yang, Jian Yang, Junyang Lin, and Chang Zhou. 2024c. Synthesizing text-to-SQL data from weak and strong LLMs. Preprint, arXiv:2408.03256. Accessed: 2025-04-16. find
  51. Bailin Wang, Richard Shin, Xiaodong Liu, Oleksandr Polozov, and Matthew Richardson. 2021. RAT-SQL: Relation-aware schema encoding and linking for text-to-SQL parsers. Preprint, arXiv:1911.04942.
  52. Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Weinan Dai, Tiantian Fan, Gaohong Liu, Lingjun Liu, and 1 others. 2025. Dapo: An open-source llm reinforcement learning system at scale. arXiv preprint arXiv:2503.14476.
  53. Junqiao Wang, Zeng Zhang, Yangfan He, Yuyang Song, Tianyu Shi, Yuchen Li, Hengyuan Xu, Kunyu Wu, Guangwu Qian, Qiuwu Chen, and Lewei He. 2025. Enhancing code LLMs with reinforcement learning in code generation: A survey. Preprint, arXiv:2412.20367. — [arXiv:2412.20367](https://arxiv.org/abs/2412.20367) find
  54. Tao Yu, Rui Zhang, Kai Yang, Michihiro Yasunaga, Dongxu Wang, Zifan Li, James Ma, Irene Li, Qingning Yao, Shanelle Roman, and 1 others. 2018. Spider: A large-scale human-labeled dataset for complex and cross-domain semantic parsing and text-to-SQL task. arXiv preprint arXiv:1809.08887.
  55. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. 2023. Chain-of-thought prompting elicits reasoning in Large Language Models. Preprint, arXiv:2201.11903.
  56. Tian Xie, Zitian Gao, Qingnan Ren, Haoming Luo, Yuqian Hong, Bryan Dai, Joey Zhou, Kai Qiu, Zhirong Wu, and Chong Luo. 2025. Logic-RL: Unleashing LLM reasoning with rule-based Reinforcement Learning. Preprint, arXiv:2502.14768.
  57. Fengli Xu, Qianyue Hao, Zefang Zong, Jingwei Wang, Yunke Zhang, Jingyi Wang, Xiaochong Lan, Jiahui Gong, Tianjian Ouyang, Fanjin Meng, Chenyang Bohan Zhai, Canwen Xu, Yuxiong He, and Zhewei Yao. 2025. ExCoT: Optimizing reasoning for text-to-SQL with execution feedback. arXiv preprint arXiv:2503.1998… find
  58. Yi Zhang, Jan Deriu, George Katsogiannis-Meimarakis, Catherine Kosten, Georgia Koutrika, and Kurt Stockinger. 2023. ScienceBenchmark: A complex real-world benchmark for evaluating natural language to SQL systems. arXiv preprint arXiv:2306.04743. — [arXiv:2306.04743](https://arxiv.org/abs/2306.04743) find
  59. Chujie Zheng, Shixuan Liu, Mingze Li, Xiong-Hui Chen, Bowen Yu, Chang Gao, Kai Dang, Yuqiong Liu, Rui Men, An Yang, and 1 others. 2025. Group sequence policy optimization. arXiv preprint arXiv:2507.18071.
  60. Victor Zhong, Caiming Xiong, and Richard Socher. 2017. Seq2SQL: Generating structured queries from natural language using reinforcement learning. Preprint, arXiv:1709.00103.
  61. Qihao Zhu, Daya Guo, Zhihong Shao, Dejian Yang, Peiyi Wang, Runxin Xu, Y Wu, Yukun Li, Huazuo Gao, Shirong Ma, and 1 others. 2024a. DeepSeek-Coder-v2: Breaking the barrier of closed-source models in code intelligence. arXiv preprint arXiv:2406.11931.
  62. Xiaohu Zhu, Qian Li, Lizhen Cui, and Yongkang Liu. 2024b. Large language model enhanced text-to-SQL generation: A survey. Preprint, arXiv:2410.06011. — [arXiv:2410.06011](https://arxiv.org/abs/2410.06011) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/yao-2026-arctic-text2sql-r1-simple-rewards/