# TRUST-SQL: Tool-Integrated Multi-Turn Reinforcement Learning for Text-to-SQL over Unknown Schemas

arXiv: https://arxiv.org/abs/2603.16448 · HTML: https://mlsanity.com/paper/2603.16448/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods.
- Builds on 25 of its 34 references that are in the map.
- First posted on arXiv 17 Mar 2026.

## Placed in

- Deep reinforcement learning methods / [RL training of language and agent models](https://mlsanity.com/deeprl/rl-training-of-language-and-agent-models/)

## Builds on (most cited first, 25 of 25)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task](https://mlsanity.com/paper/1809.08887/) (2018)
- [Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs](https://mlsanity.com/paper/2305.03111/) (2023)
- [LLMs Get Lost In Multi-Turn Conversation](https://mlsanity.com/paper/2505.06120/) (2025)
- [RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning](https://mlsanity.com/paper/2504.20073/) (2025)
- [CHESS: Contextual Harnessing for Efficient SQL Synthesis](https://mlsanity.com/paper/2405.16755/) (2024)
- [MAC-SQL: A Multi-Agent Collaborative Framework for Text-to-SQL](https://mlsanity.com/paper/2312.11242/) (2023)
- [SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning](https://mlsanity.com/paper/2509.02479/) (2025)
- [Structure-Grounded Pretraining for Text-to-SQL](https://mlsanity.com/paper/2010.12773/) (2020)
- [Towards Robustness of Text-to-SQL Models against Synonym Substitution](https://mlsanity.com/paper/2106.01065/) (2021)
- [Exploring Underexplored Limitations of Cross-Domain Text-to-SQL Generalization](https://mlsanity.com/paper/2109.05157/) (2021)
- [SQL-R1: Training Natural Language to SQL Reasoning Model By Reinforcement Learning](https://mlsanity.com/paper/2504.08600/) (2025)
- [Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL](https://mlsanity.com/paper/2505.20315/) (2025)
- [Enhancing Personalized Multi-Turn Dialogue with Curiosity Reward](https://mlsanity.com/paper/2504.03206/) (2025)
- [Automatic Metadata Extraction for Text-to-SQL](https://mlsanity.com/paper/2505.19988/) (2025)
- [Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scaling](https://mlsanity.com/paper/2509.24403/) (2025)
- [STaR-SQL: Self-Taught Reasoner for Text-to-SQL](https://mlsanity.com/paper/2502.13550/) (2025)
- [AutoLink: Autonomous Schema Exploration and Expansion for Scalable Schema Linking in Text-to-SQL at Scale](https://mlsanity.com/paper/2511.17190/) (2025)
- [MTIR-SQL: Multi-turn Tool-Integrated Reasoning Reinforcement Learning for Text-to-SQL](https://mlsanity.com/paper/2510.25510/) (2025)
- [SQL-Trail: Multi-Turn Reinforcement Learning with Interleaved Feedback for Text-to-SQL](https://mlsanity.com/paper/2601.17699/) (2026)
- [MTSQL-R1: Towards Long-Horizon Multi-Turn Text-to-SQL via Agentic Training](https://mlsanity.com/paper/2510.12831/) (2025)
- [EvoSchema: Towards Text-to-SQL Robustness Against Schema Evolution](https://mlsanity.com/paper/2603.10697/) (2026)
- [AGRO-SQL: Agentic Group-Relative Optimization with High-Fidelity Data Synthesis](https://mlsanity.com/paper/2512.23366/) (2025)
- [Harmonizing Dense and Sparse Signals in Multi-turn RL: Dual-Horizon Credit Assignment for Industrial Sales Agents](https://mlsanity.com/paper/2603.01481/) (2026)

## Built on by (most cited first, 3 of 3)

- [ACE-SQL: Adaptive Co-Optimization via Empirical Credit Assignment for Text-to-SQL](https://mlsanity.com/paper/2606.05906/) (2026)
- [SERL-SQL: Selective Hindsight Distillation for Text-to-SQL Reinforcement Agentic Learning](https://mlsanity.com/paper/2608.00485/) (2026)
- [C3: Rethinking Process Supervision via Confidence-Guided Credit Correction in Multi-Turn Reinforcement Learning](https://mlsanity.com/paper/2610.19695/) (2026)

## Abstract (arXiv)

Text-to-SQL parsing has achieved remarkable progress under the Full Schema Assumption. However, this premise fails in real-world enterprise environments where databases contain hundreds of tables with massive noisy metadata. Rather than injecting the full schema upfront, an agent must actively identify and verify only the relevant subset, giving rise to the Unknown Schema scenario we study in this work. To address this, we propose TRUST-SQL (Truthful Reasoning with Unknown Schema via Tools). We formulate the task as a Partially Observable Markov Decision Process where our autonomous agent employs a structured four-phase protocol to ground reasoning in verified metadata. Crucially, this protocol provides a structural boundary for our novel Dual-Track GRPO strategy. By applying token-level masked advantages, this strategy isolates exploration rewards from execution outcomes to resolve credit assignment, yielding a 9.9% relative improvement over standard GRPO. Extensive experiments across five benchmarks demonstrate that TRUST-SQL achieves an average absolute improvement of 30.6% and 16.6% for the 4B and 8B variants respectively over their base models. Remarkably, despite operating entirely without pre-loaded metadata, our framework consistently matches or surpasses strong baselines that rely on schema prefilling. Data, code, and checkpoints are available at https://huggingface.co/collections/AIJian/trustsql.