# ConstrainedSQL: Training LLMs for Text2SQL via Constrained Reinforcement Learning

arXiv: https://arxiv.org/abs/2511.09693 · HTML: https://mlsanity.com/paper/2511.09693/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- First posted on arXiv 12 Nov 2025.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)

## Built on by (most cited first, 1 of 1)

- [Text-to-SPARQL Generation with Reinforcement Learning: A GRPO-based Approach on DBLP](https://mlsanity.com/paper/2605.20066/) (2026)

## Abstract (arXiv)

Reinforcement learning (RL) has demonstrated significant promise in enhancing the reasoning capabilities of Text2SQL LLMs, especially with advanced algorithms such as GRPO and DAPO. However, the performance of these methods is highly sensitive to the design of reward functions. Inappropriate rewards can lead to reward hacking, where models exploit loopholes in the reward structure to achieve high scores without genuinely solving the task. This work considers a constrained RL framework for Text2SQL that incorporates natural and interpretable reward and constraint signals, while dynamically balancing trade-offs among them during the training. We establish the theoretical guarantees of our constrained RL framework and our numerical experiments on the well-known Text2SQL datasets substantiate the improvement of our approach over the state-of-the-art RL-trained LLMs.