Progress-SQL: Improving Reinforcement Learning for Text-to-SQL via Progressive Rewards

2026-06-05 · Shihao Zhang, Xiaoman Wang, Yuan Liu, Yunshi Lan, Weining Qian arxiv

Reinforcement learning has recently shown promise in improving large language models for Text-to-SQL generation, yet existing methods typically optimize one-shot rewards defined over a single SQL state. Such rewards provide limited guidance for iterative SQL correction and are insufficient to capture the improvement of multi-turn SQL refinement. In this paper, we propose Progress-SQL, a multi-turn reinforcement learning framework with progressive rewards for Text-to-SQL. Our approach introduces an Oracle-guided Diagnostic Tree (ODT), which abstracts SQL queries into clause-level structural profiles and produces diagnostic feedback for next-turn refinement. To provide dense and robust reward signals, we combine ODT-based structural alignment with lexical alignment and define a progressive reward that measures the improvement from the initial SQL to the final SQL. We further incorporate a progression latency reward that favors earlier correctness and an execution status reward that encourages recovery from the invalid SQL. Experiments on BIRD, Spider, and Spider robustness variants demonstrate that our method consistently improves Text-to-SQL performance across both primary and robustness evaluations.

📄 PDF Abstract BibTeX arXiv:2606.06825

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Non-vacuous Generalization Bounds for Reinforcement Learning with Verifiable Rewards

2026-07-16 · Yuxuan Zhu, Rohan Alur, Daniel Kang arxiv

While reinforcement learning with verifiable rewards (RLVR) is widely used to improve the reasoning capabilities of large language models (LLMs), the generalizability of the resulting models remains poorly understood. In…

Reinforcement Learning

Tree-Structured Policy based Progressive Reinforcement Learning for Temporally Language Grounding in Video

2020-01-18 · Jie Wu, Guanbin Li, Si Liu, Liang Lin

Temporally language grounding in untrimmed videos is a newly-raised task in video understanding. Most of the existing methods suffer from inferior efficiency, lacking interpretability, and deviating from the human percep…

Decision Making reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)+1

Adapting to Reward Progressivity via Spectral Reinforcement Learning

2021-04-29 · ICLR 2021 1 · Michael Dann, John Thangarajah

In this paper we consider reinforcement learning tasks with progressive rewards; that is, tasks where the rewards tend to increase in magnitude over time. We hypothesise that this property may be problematic for value-ba…

Atari Games Deep Reinforcement Learning reinforcement-learning Reinforcement Learning+1

Progressive Reasoning with Primitive Correction for Compositional Zero-Shot Learning

2026-07-07 · Ziyi Chen, Haoyan Shi, Sunhan Xu, Congyan Lang arxiv

Compositional Zero-Shot Learning (CZSL) aims to combine known attributes and objects as primitives for recognizing previously unseen attribute-object pairs. Prior works either predict attributes and objects independently…

Compositional Zero-Shot Learning Reinforcement Learning

Sim-to-Real Robot Learning from Pixels with Progressive Nets

2016-10-13 · Andrei A. Rusu, Mel Vecerik, Thomas Rothörl, Nicolas Heess 외

Applying end-to-end learning to solve complex, interactive, pixel-driven control tasks on a robot is an unsolved problem. Deep Reinforcement Learning algorithms are too slow to achieve performance on a real robot, but th…

Deep Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)+1