PaVeRL-SQL: Text–to–SQL via Partial–Match Rewards and Verbal Reinforcement Learning
Authors:
- 8 December 2025
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 7687-7696
Abstract
Text–to–SQL models allow users to interact with a database more easily by generating executable SQL statements from natural–language questions. Despite recent successes on simpler databases and questions, current Text–to–SQL methods still suffer from low execution accuracy on industry–scale databases and complex questions involving domain–specific business logic. We present PaVeRL–SQL, a framework that combines Partial–Match Rewards and Verbal Reinforcement Learning to drive self–improvement in reasoning language models (RLMs) for Text–to–SQL. To handle practical use cases, we adopt two pipelines: (1) a newly designed in–context learning framework with group self–evaluation (verbal–RL), using capable open–and closed–source large language models (LLMs) as backbones; and (2) a chain–of–thought (CoT) RL pipeline with a small backbone model (OmniSQL–7B) trained with a specially designed reward function and two–stage RL. These pipelines achieve state–of–the–art (SOTA) results on popular Text–to–SQL bench-marks–Spider, Spider 2.0, and BIRD. For the industrial–level Spider2.0–SQLite benchmark, the verbal–RL pipeline achieves an execution accuracy 7.4% higher than SOTA, and the CoT pipeline is 1.4% higher. RL training with mixed SQL dialects yields strong, threefold gains, particularly for dialects with limited training data. Overall, PaVeRL–SQL delivers reliable, SOTA Text–to–SQL under realistic industrial constraints. The code is available at https://github.com/PaVeRL-SQL/PaVeRL-SQL.Keywords
This publication has 10 references indexed in Scilit:
- Demystifying LLM-Based Software Engineering AgentsProceedings of the ACM on Software Engineering, 2025
- HybridFlow: A Flexible and Efficient RLHF Framework2025
- STaR-SQL: Self-Taught Reasoner for Text-to-SQL2025
- CodeS: Towards Building Open-source Language Models for Text-to-SQLProceedings of the ACM on Management of Data, 2024
- RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQLProceedings of the AAAI Conference on Artificial Intelligence, 2023
- ACT-SQL: In-Context Learning for Text-to-SQL with Automatically-Generated Chain-of-Thought2023
- CRUSH4SQL: Collective Retrieval Using Schema Hallucination For Text2SQL2023
- Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models2023
- PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models2021
- RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers2020