mlsanityData to 14 Oct 2026

DualSQL: Text-to-SQL with Multi-Agent Reinforcement Learning

arXiv 2609.18135 ↗First posted 16 Sep 2026cs.CL cs.AI cs.DBcited by 0

Not on the map. Why?

Abstract (arXiv)

State-of-the-art Text-to-SQL systems are typically multi-agent pipelines centered around two fundamental tasks: schema linking and SQL generation. However, existing work trains separate models for each task, failing to leverage the synergy between these interrelated tasks. In this work, we propose DualSQL, a new Text-to-SQL system consisting of two agents powered by a single model backbone. The agents share the same model weights and agentic scaffold, enabling joint optimization through a robust multi-agent reinforcement learning (RL) framework. We design three database access tools to facilitate effective multi-step reasoning grounded to interactions with the databases. To improve training and avoid model collapse, we introduce a set of rollout guardrail mechanisms that stabilizes multi-agent RL training, supporting DualSQL to keep improving during training. We also introduce a new SQL correctness metric, robust execution match (REX), to more accurately judge SQL correctness and assign reward signals. Being trained on only 3755 examples, DualSQL-4B achieves an impressive 68.0% execution accuracy on the BIRD development set, matching previous 7B models. DualSQL-8B further improves to 71.1%, outperforming previous state-of-the-art single-model solutions with 32B parameters. These results demonstrate the strength of joint multi-agent reinforcement learning for building high performance Text-to-SQL pipelines.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 33 of its 42 references that are in the map. First posted on arXiv 16 Sep 2026.

Builds on

33 of 42 references.

References · 33

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  3. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
  4. 2018Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Taskcited by 1,139
  5. 2023Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLscited by 688
  6. 2025Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoningcited by 559
  7. 2023DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correctioncited by 477
  8. 2023Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluationcited by 385
  9. 2024CodeS: Towards Building Open-source Language Models for Text-to-SQLcited by 187
  10. 2024CHESS: Contextual Harnessing for Efficient SQL Synthesiscited by 169
  11. 2023MAC-SQL: A Multi-Agent Collaborative Framework for Text-to-SQLcited by 164
  12. 2025SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoningcited by 146
21 more
  1. 2024CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQLcited by 138
  2. 2025Context Length Alone Hurts LLM Performance Despite Perfect Retrievalcited by 127
  3. 2025OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scalecited by 91
  4. 2025Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learningcited by 84
  5. 2024TableGPT2: A Large Multimodal Model with Tabular Data Integrationcited by 72
  6. 2025SQL-R1: Training Natural Language to SQL Reasoning Model By Reinforcement Learningcited by 68
  7. 2025Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQLcited by 59
  8. 2025LongReason: A Synthetic Long-Context Reasoning Benchmark via Context Expansioncited by 50
  9. 2024Understanding the Dark Side of LLMs' Intrinsic Self-Correctioncited by 42
  10. 2025Stabilizing Reinforcement Learning with LLMs: Formulation and Practicescited by 41
  11. 2025XiYan-SQL: A Novel Multi-Generator Framework For Text-to-SQLcited by 40
  12. 2025Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQLcited by 38
  13. 2025SFR-DeepResearch: Towards Effective Reinforcement Learning for Autonomously Reasoning Single Agentscited by 31
  14. 2025Is Long Context All You Need? Leveraging LLM's Extended Context for NL2SQLcited by 30
  15. 2025Automatic Metadata Extraction for Text-to-SQLcited by 27
  16. 2025Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scalingcited by 24
  17. 2025MARS-SQL: A multi-agent reinforcement learning framework for Text-to-SQLcited by 15
  18. 2025Evaluating Long-Context Reasoning in LLM-Based WebAgentscited by 15
  19. 2026AgentSM: Semantic Memory for Agentic Text-to-SQLcited by 14
  20. 2025MTIR-SQL: Multi-turn Tool-Integrated Reasoning Reinforcement Learning for Text-to-SQLcited by 10
  21. 2025DB-Explore: Automated Database Exploration and Instruction Synthesis for Text-to-SQLcited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026ReToolSQL: Agentic Reinforcement Learning for Robust Text-to-SQL
  2. 2026SQL-Trail: Multi-Turn Reinforcement Learning with Interleaved Feedback for Text-to-SQL
  3. 2025A Multi-agent Text2SQL Framework using Small Language Models and Execution Feedback
  4. 2025CogniSQL-R1-Zero: Lightweight Reinforced Reasoning for Efficient SQL Generation
  5. 2025ConstrainedSQL: Training LLMs for Text2SQL via Constrained Reinforcement Learning
  6. 2026HarnessSQL: Harness-Native Training for SQL Agents in Realistic Database Environments
  7. 2025Sparks of Tabular Reasoning via Text2SQL Reinforcement Learning
  8. 2025PaVeRL-SQL: Text-to-SQL via Partial-Match Rewards and Verbal Reinforcement Learning