Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2509.07159 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916940495192064 |
|---|---|
| author | Hao, Heng Hu, Wenjun Verkholyak, Oxana Tarzanagh, Davoud Ataee Gutow, Baruch Didari, Sima Faraki, Masoud Moon, Hankyu Min, Seungjai |
| author_facet | Hao, Heng Hu, Wenjun Verkholyak, Oxana Tarzanagh, Davoud Ataee Gutow, Baruch Didari, Sima Faraki, Masoud Moon, Hankyu Min, Seungjai |
| contents | 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 \emph{PaVeRL-SQL}, a framework that combines \emph{Partial-Match Rewards} and \emph{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 benchmarks -- 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, \emph{PaVeRL-SQL} delivers reliable, SOTA Text-to-SQL under realistic industrial constraints. The code is available at https://github.com/PaVeRL-SQL/PaVeRL-SQL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_07159 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PaVeRL-SQL: Text-to-SQL via Partial-Match Rewards and Verbal Reinforcement Learning Hao, Heng Hu, Wenjun Verkholyak, Oxana Tarzanagh, Davoud Ataee Gutow, Baruch Didari, Sima Faraki, Masoud Moon, Hankyu Min, Seungjai Artificial Intelligence 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 \emph{PaVeRL-SQL}, a framework that combines \emph{Partial-Match Rewards} and \emph{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 benchmarks -- 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, \emph{PaVeRL-SQL} delivers reliable, SOTA Text-to-SQL under realistic industrial constraints. The code is available at https://github.com/PaVeRL-SQL/PaVeRL-SQL. |
| title | PaVeRL-SQL: Text-to-SQL via Partial-Match Rewards and Verbal Reinforcement Learning |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2509.07159 |