RetrySQL: text-to-SQL training with retry data for self-correcting query generation

• URL: http://arxiv.org/abs/2507.02529v1
• Date: Thu, 03 Jul 2025 11:00:49 GMT
• Title: RetrySQL: text-to-SQL training with retry data for self-correcting query generation
• Authors: Alicja Rączkowska, Riccardo Belluzzo, Piotr Zieliński, Joanna Baran, Paweł Olszewski,
• Abstract summary: We introduce Retry, a new approach to training text-to-generation models.<n>We demonstrate that retry steps yield an improvement of up to 4 percentage points in both overall and challenging execution accuracy metrics.
• Score: 1.6707278580444538
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The text-to-SQL task is an active challenge in Natural Language Processing. Many existing solutions focus on using black-box language models extended with specialized components within customized end-to-end text-to-SQL pipelines. While these solutions use both closed-source proprietary language models and coding-oriented open-source models, there is a lack of research regarding SQL-specific generative models. At the same time, recent advancements in self-correcting generation strategies show promise for improving the capabilities of existing architectures. The application of these concepts to the text-to-SQL task remains unexplored. In this paper, we introduce RetrySQL, a new approach to training text-to-SQL generation models. We prepare reasoning steps for reference SQL queries and then corrupt them to create retry data that contains both incorrect and corrected steps, divided with a special token. We continuously pre-train an open-source coding model with this data and demonstrate that retry steps yield an improvement of up to 4 percentage points in both overall and challenging execution accuracy metrics, compared to pre-training without retry data. Additionally, we confirm that supervised fine-tuning with LoRA is ineffective for learning from retry data and that full-parameter pre-training is a necessary requirement for that task. We showcase that the self-correcting behavior is learned by the model and the increase in downstream accuracy metrics is a result of this additional skill. Finally, we incorporate RetrySQL-trained models into the full text-to-SQL pipeline and showcase that they are competitive in terms of execution accuracy with proprietary models that contain orders of magnitude more parameters. RetrySQL demonstrates that self-correction can be learned in the text-to-SQL task and provides a novel way of improving generation accuracy for SQL-oriented language models.

Related papers

• ExeSQL: Self-Taught Text-to-SQL Models with Execution-Driven Bootstrapping for SQL Dialects [24.450818792474216]
  This work introduces Exe, a text-to-guided framework with execution-driven, agentic bootstrapping.<n>We show that Exe bridges the dialect gap in text-to-guided learning, achieving average improvements of 15.2%, 10.38%, and 4.49% over GPT-4o on, and Oracle, respectively.
  arXiv  Detail & Related papers  (2025-05-22T19:13:34Z)
• STaR-SQL: Self-Taught Reasoner for Text-to-SQL [20.719165038519744]
  "chain-of-thought" rationales have proven effective for improving the performance of large language models on complex reasoning tasks.<n>Applying such techniques to structured tasks, such as text-to-driven, remains largely unexplored.<n>In this paper, we introduce Self-Taughter for text-to-driven (STaR-), a novel approach that reframes query generation as a reasoning process.<n> Experimental results on the challenging Spider benchmark demonstrate that STaR- significantly improves text-to-performance, achieving an execution accuracy of 86.6%.<n>These findings underscore the potential of reasoning-augmented training for
  arXiv  Detail & Related papers  (2025-02-19T08:58:44Z)
• Context-Aware SQL Error Correction Using Few-Shot Learning -- A Novel Approach Based on NLQ, Error, and SQL Similarity [0.0]
  This paper introduces a novel few-shot learning-based approach for error correction insql generation. It enhances the accuracy of generated queries by selecting the most suitable few-shot error correction examples for a given natural language question (NLQ) In experiments with the open-source dataset, the proposed model offers a 39.2% increase in fixing errors with no error correction and a 10% increase from a simple error correction method.
  arXiv  Detail & Related papers  (2024-10-11T18:22:08Z)
• SQL-GEN: Bridging the Dialect Gap for Text-to-SQL Via Synthetic Data And Model Merging [30.306023265985658]
  We introduce a framework for generating high-quality synthetic training data for any dialect. We propose a novel Mixture-of-Experts (MoE) that leverages the shared knowledge across dialects.
  arXiv  Detail & Related papers  (2024-08-22T20:50:48Z)
• DAC: Decomposed Automation Correction for Text-to-SQL [51.48239006107272]
  We introduce De Automation Correction (DAC), which corrects text-to-composed by decomposing entity linking and skeleton parsing. We show that our method improves performance by $3.7%$ on average of Spider, Bird, and KaggleDBQA compared with the baseline method.
  arXiv  Detail & Related papers  (2024-08-16T14:43:15Z)
• SQL-PaLM: Improved Large Language Model Adaptation for Text-to-SQL (extended) [53.95151604061761]
  This paper introduces the framework for enhancing Text-to- filtering using large language models (LLMs) With few-shot prompting, we explore the effectiveness of consistency decoding with execution-based error analyses. With instruction fine-tuning, we delve deep in understanding the critical paradigms that influence the performance of tuned LLMs.
  arXiv  Detail & Related papers  (2023-05-26T21:39:05Z)
• Wav2SQL: Direct Generalizable Speech-To-SQL Parsing [55.10009651476589]
  Speech-to-Spider (S2Spider) aims to convert spoken questions intosql queries given databases. We propose the first direct speech-to-speaker parsing model Wav2 which avoids error compounding across cascaded systems. Experimental results demonstrate that Wav2 avoids error compounding and achieves state-of-the-art results by up to 2.5% accuracy improvement over the baseline.
  arXiv  Detail & Related papers  (2023-05-21T19:26:46Z)
• SPSQL: Step-by-step Parsing Based Framework for Text-to-SQL Generation [13.196264569882777]
  The current mainstream end-to-end Text2 model is not only difficult to build due to its complex structure and high requirements for training data, but also difficult to adjust due to massive parameters. This paper proposes a pipeline method: SP Experiments to achieve the desired result. We construct the dataset based on the marketing business data of the State Grid Corporation of China.
  arXiv  Detail & Related papers  (2023-05-10T10:01:36Z)
• Augmenting Multi-Turn Text-to-SQL Datasets with Self-Play [46.07002748587857]
  We explore augmenting the training datasets using self-play, which leverages contextual information to synthesize new interactions. We find that self-play improves the accuracy of a strong baseline on SParC and Co, two widely used text-to-domain datasets.
  arXiv  Detail & Related papers  (2022-10-21T16:40:07Z)
• A Survey on Text-to-SQL Parsing: Concepts, Methods, and Future Directions [102.8606542189429]
  The goal of text-to-corpora parsing is to convert a natural language (NL) question to its corresponding structured query language () based on the evidences provided by databases. Deep neural networks have significantly advanced this task by neural generation models, which automatically learn a mapping function from an input NL question to an output query.
  arXiv  Detail & Related papers  (2022-08-29T14:24:13Z)
• Weakly Supervised Text-to-SQL Parsing through Question Decomposition [53.22128541030441]
  We take advantage of the recently proposed question meaning representation called QDMR. Given questions, their QDMR structures (annotated by non-experts or automatically predicted) and the answers, we are able to automatically synthesizesql queries. Our results show that the weakly supervised models perform competitively with those trained on NL- benchmark data.
  arXiv  Detail & Related papers  (2021-12-12T20:02:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.