mlsanityData to 14 Oct 2026

Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL

arXiv 2503.23157 ↗First posted 29 Mar 2025cs.LG cs.AI cs.DBcited by 59

On the map: LLM reasoning → Knowledge- and tool-augmented reasoning

Abstract (arXiv)

Text-to-SQL is a challenging task involving multiple reasoning-intensive subtasks, including natural language understanding, database schema comprehension, and precise SQL query formulation. Existing approaches often rely on handcrafted reasoning paths with inductive biases that can limit their overall effectiveness. Motivated by the recent success of reasoning-enhanced models such as DeepSeek R1 and OpenAI o1, which effectively leverage reward-driven self-exploration to enhance reasoning capabilities and generalization, we propose a novel set of partial rewards tailored specifically for the Text-to-SQL task. Our reward set includes schema-linking, AI feedback, n-gram similarity, and syntax check, explicitly designed to address the reward sparsity issue prevalent in reinforcement learning (RL). Leveraging group relative policy optimization (GRPO), our approach explicitly encourages large language models (LLMs) to develop intrinsic reasoning skills necessary for accurate SQL query generation. With models of different sizes, we demonstrate that RL-only training with our proposed rewards consistently achieves higher accuracy and superior generalization compared to supervised fine-tuning (SFT). Remarkably, our RL-trained 14B-parameter model significantly outperforms larger proprietary models, e.g. o3-mini by 4% and Gemini-1.5-Pro-002 by 3% on the BIRD benchmark. These highlight the efficacy of our proposed RL-training framework with partial rewards for enhancing both accuracy and reasoning capabilities in Text-to-SQL tasks.

Key facts. Cited by 59 papers in the map (data through 14 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 50 of its 60 references that are in the map. First posted on arXiv 29 Mar 2025.

Builds on

50 of 60 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQL2 Oct
  2. 2024The Death of Schema Linking? Text-to-SQL in the Age of Well-Reasoned Language Models14 Aug
  3. 2023ACT-SQL: In-Context Learning for Text-to-SQL with Automatically-Generated Chain-of-Thought26 Oct

Other references · 47

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2017Proximal Policy Optimization Algorithmscited by 15,215
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  4. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  6. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  7. 2023Let's Verify Step by Stepcited by 3,789
  8. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  9. 2025s1: Simple test-time scalingcited by 1,341
  10. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  11. 2024Qwen2.5-Coder Technical Reportcited by 1,203
  12. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 940
35 more
  1. 2023Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLscited by 688
  2. 2017Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learningcited by 687
  3. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  4. 2023DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correctioncited by 477
  5. 2019RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parserscited by 424
  6. 2021PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Modelscited by 411
  7. 2016Reinforcement Learning with Unsupervised Auxiliary Taskscited by 404
  8. 2023Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluationcited by 385
  9. 2025SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolutioncited by 263
  10. 2019Towards Complex Text-to-SQL in Cross-Domain Database with Intermediate Representationcited by 210
  11. 2024CodeS: Towards Building Open-source Language Models for Text-to-SQLcited by 187
  12. 2024CHESS: Contextual Harnessing for Efficient SQL Synthesiscited by 169
  13. 2023C3: Zero-shot Text-to-SQL with ChatGPTcited by 164
  14. 2024Marco-o1: Towards Open Reasoning Models for Open-Ended Solutionscited by 144
  15. 2024O1 Replication Journey: A Strategic Progress Report -- Part 1cited by 143
  16. 2017SQLNet: Generating Structured Queries From Natural Language Without Reinforcement Learningcited by 135
  17. 2021LGESQL: Line Graph Enhanced Text-to-SQL Model with Mixed Local and Non-Local Relationscited by 119
  18. 2021Towards Robustness of Text-to-SQL Models against Synonym Substitutioncited by 110
  19. 2024MCS-SQL: Leveraging Multiple Prompts and Multiple-Choice Selection For Text-to-SQL Generationcited by 105
  20. 2025ACECODER: Acing Coder RL via Automated Test-Case Synthesiscited by 93
  21. 2021Exploring Underexplored Limitations of Cross-Domain Text-to-SQL Generalizationcited by 92
  22. 2024The Dawn of Natural Language to SQL: Are We Fully Ready?cited by 79
  23. 2022Reinforcement Learning with Sparse Rewards using Guidance from Offline Demonstrationcited by 54
  24. 2025OpenSearch-SQL: Enhancing Text-to-SQL with Dynamic Few-shot and Consistency Alignmentcited by 53
  25. 2024A Preview of XiYan-SQL: A Multi-Generator Ensemble Framework for Text-to-SQLcited by 46
  26. 2023Enhancing Few-shot Text-to-SQL Capabilities of Large Language Models: A Study on Prompt Design Strategiescited by 45
  27. 2024DTS-SQL: Decomposed Text-to-SQL with Small Large Language Modelscited by 45
  28. 2021SADGA: Structure-Aware Dual Graph Aggregation Network for Text-to-SQLcited by 43
  29. 2023SQL-PaLM: Improved Large Language Model Adaptation for Text-to-SQL (extended)cited by 42
  30. 2024E-SQL: Direct Schema Linking via Question Enrichment in Text-to-SQLcited by 35
  31. 2025AlphaMaze: Enhancing Large Language Models' Spatial Intelligence via GRPOcited by 33
  32. 2022Reward Design For An Online Reinforcement Learning Algorithm Supporting Oral Self-Carecited by 18
  33. 2020Reinforcement Learning Approaches in Social Roboticscited by 16
  34. 2023SQLPrompt: In-Context Text-to-SQL with Minimal Labeled Datacited by 16
  35. 2021Unbiased Methods for Multi-Goal Reinforcement Learningcited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 34; 2026: 25.

59 papers.

Sort:

By area:

  1. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 180
  2. 2025VerlTool: Towards Holistic Agentic Reinforcement Learning with Tool Usecited by 96
  3. 2025Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQLcited by 38
  4. 2025Reward-SQL: Boosting Text-to-SQL via Stepwise Execution-Aware Reasoning and Process-Supervised Rewardscited by 27
  5. 2025Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scalingcited by 24
  6. 2025BIRD-INTERACT: Re-imagining Text-to-SQL Evaluation for Large Language Models via Lens of Dynamic Interactionscited by 19
  7. 2025Reasoning-Table: Exploring Reinforcement Learning for Table Reasoningcited by 18
  8. 2025AutoLink: Autonomous Schema Exploration and Expansion for Scalable Schema Linking in Text-to-SQL at Scalecited by 17
  9. 2025CSC-SQL: Corrective Self-Consistency in Text-to-SQL via Reinforcement Learningcited by 15
  10. 2025MARS-SQL: A multi-agent reinforcement learning framework for Text-to-SQLcited by 15
  11. 2025Think2SQL: Reinforce LLM Reasoning Capabilities for Text2SQLcited by 14
  12. 2025Cheaper, Better, Faster, Stronger: Robust Text-to-SQL without Chain-of-Thought or Fine-Tuningcited by 13
38 more
  1. 2026Pervasive Annotation Errors Break Text-to-SQL Benchmarks and Leaderboardscited by 13
  2. 2025Co-PatcheR: Collaborative Software Patching with Component(s)-specific Small Reasoning Modelscited by 12
  3. 2025SQLens: An End-to-End Framework for Error Detection and Correction in Text-to-SQLcited by 12
  4. 2025Text2SQL-Flow: A Robust SQL-Aware Data Augmentation Framework for Text-to-SQLcited by 11
  5. 2025SLM-SQL: An Exploration of Small Language Models for Text-to-SQLcited by 10
  6. 2025MTIR-SQL: Multi-turn Tool-Integrated Reasoning Reinforcement Learning for Text-to-SQLcited by 10
  7. 2025Sparks of Tabular Reasoning via Text2SQL Reinforcement Learningcited by 9
  8. 2025LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Modelscited by 8
  9. 2025Graph-Reward-SQL: Execution-Free Reinforcement Learning for Text-to-SQL via Graph Matching and Stepwise Rewardcited by 7
  10. 2025LogicCat: A Chain-of-Thought Text-to-SQL Benchmark for Complex Reasoningcited by 7
  11. 2025STRuCT-LLM: Unifying Tabular and Graph Reasoning with Reinforcement Learning for Semantic Parsingcited by 5
  12. 2025SING-SQL: A Synthetic Data Generation Framework for In-Domain Text-to-SQL Translationcited by 5
  13. 2025MTSQL-R1: Towards Long-Horizon Multi-Turn Text-to-SQL via Agentic Trainingcited by 5
  14. 2025RAISE: Reasoning Agent for Interactive SQL Explorationcited by 4
  15. 2026RoboPhD: Self-Improving Text-to-SQL Through Autonomous Agent Evolutioncited by 3
  16. 2026Closing the Reflection Gap: A Free Calibration Bonus for Agentic RLcited by 3
  17. 2025CogniSQL-R1-Zero: Lightweight Reinforced Reasoning for Efficient SQL Generationcited by 2
  18. 2025Toward Real-World Table Agents: Capabilities, Workflows, and Design Principles for LLM-based Table Intelligencecited by 2
  19. 2025PaVeRL-SQL: Text-to-SQL via Partial-Match Rewards and Verbal Reinforcement Learningcited by 2
  20. 2025Rethinking Text-to-SQL: Dynamic Multi-turn SQL Interaction for Real-world Database Explorationcited by 2
  21. 2025Thinking by Doing: Building Efficient World Model Reasoning in LLMs via Multi-turn Interactioncited by 2
  22. 2025AGRO-SQL: Agentic Group-Relative Optimization with High-Fidelity Data Synthesiscited by 2
  23. 2025HES-SQL: Hybrid Reasoning for Efficient Text-to-SQL with Structural Skeleton Guidancecited by 1
  24. 2025LitE-SQL: A Lightweight and Efficient Text-to-SQL Framework with Vector-based Schema Linking and Execution-Guided Self-Correctioncited by 1
  25. 2026EvolSQL: Structure-Aware Evolution for Scalable Text-to-SQL Data Synthesiscited by 1
  26. 2026Beyond Static Pipelines: Learning Dynamic Workflows for Text-to-SQLcited by 1
  27. 2026Human-Level Text-to-SQL via Reinforcement Learning on Verified Data, Without Pipeline Engineeringcited by 1
  28. 2026NL2SQLBench: A Modular Benchmarking Framework for LLM-Enabled NL2SQL Solutionscited by 1
  29. 2026Every Step Counts: Step-Level Credit Assignment for Tool-Integrated Text-to-SQLcited by 1
  30. 2026Text-to-SPARQL Generation with Reinforcement Learning: A GRPO-based Approach on DBLPcited by 1
  31. 2026TAHOE: Text-to-SQL with Automated Hint Optimization from Experiencecited by 1
  32. 2026SQLConductor: Search-to-Policy Learning for Step-wise Text-to-SQL Orchestrationcited by 1
  33. 2026SQLMorph: Query Mutation and Fine-Grained Metrics for Text-to-SQL Evaluationcited by 1
  34. 2025BAPPA: Benchmarking Agents, Plans, and Pipelines for Automated Text-to-SQL Generationcited by 0
  35. 2026RingSQL: Schema-Independent Synthetic Data Generation for Text-to-SQL Reinforcement Learningcited by 0
  36. 2026FINER-SQL: Boosting Small Language Models for Text-to-SQLcited by 0
  37. 2026Residual Skill Optimization for Text-to-SQL Ensemblescited by 0
  38. 2026SIRIUS-SQL: Anchoring Multi-Candidate Text-to-SQL in Execution Feedbackcited by 0

50 most cited of 59 All 59 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Progress-SQL: Improving Reinforcement Learning for Text-to-SQL via Progressive Rewards
  2. 2025Beyond Query-Level Comparison: Fine-Grained Reinforcement Learning for Text-to-SQL with Automated Interpretable Critiques
  3. 2025ReEx-SQL: Reasoning with Execution-Aware Reinforcement Learning for Text-to-SQL
  4. 2026SQL-Trail: Multi-Turn Reinforcement Learning with Interleaved Feedback for Text-to-SQL
  5. 2025JudgeSQL: Reasoning over SQL Candidates with Weighted Consensus Tournament
  6. 2025ConstrainedSQL: Training LLMs for Text2SQL via Constrained Reinforcement Learning
  7. 2025Schema-R1: A reasoning training approach for schema linking in Text-to-SQL Task
  8. 2026ReToolSQL: Agentic Reinforcement Learning for Robust Text-to-SQL