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

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

作者：Shijie Chen, Yu Gan, Yeounoh Chung, Jiani Zhang, Quannan Li, Sravan Babu Bodapati, Cody J. Greer, Yu Su, Fatma Ozcan

来源：arXiv cs.DB（数据库）

正文

Computer Science > Computation and Language

arXiv:2609.18135v1 (cs)

[Submitted on 16 Sep 2026]

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

Authors:Shijie Chen, Yu Gan, Yeounoh Chung, Jiani Zhang, Quannan Li, Sravan Babu Bodapati, Cody J. Greer, Yu Su, Fatma Ozcan

View a PDF of the paper titled DualSQL: Text-to-SQL with Multi-Agent Reinforcement Learning, by Shijie Chen and 8 other authors

View PDF

HTML (experimental)

Abstract: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.

Subjects:

Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Databases (cs.DB)

Cite as:

arXiv:2609.18135 [cs.CL]

(or

arXiv:2609.18135v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2609.18135

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

主题

自然语言处理 · 人工智能 · 数据库

由「前沿雷达」于 2026-09-20 采集。正文取自原文页面，已保留出处链接。标题与正文版权归原作者所有。