Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Jinyang, Li, Xiaolong, Qu, Ge, Jacobsson, Per, Qin, Bowen, Hui, Binyuan, Si, Shuzheng, Huo, Nan, Xu, Xiaohan, Zhang, Yue, Tang, Ziwei, Li, Yuanshuai, Widjaja, Florensia, Zhu, Xintong, Zhou, Feige, Huang, Yongfeng, Papakonstantinou, Yannis, Ozcan, Fatma, Ma, Chenhao, Cheng, Reynold
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2506.18951
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917219943841792
author Li, Jinyang
Li, Xiaolong
Qu, Ge
Jacobsson, Per
Qin, Bowen
Hui, Binyuan
Si, Shuzheng
Huo, Nan
Xu, Xiaohan
Zhang, Yue
Tang, Ziwei
Li, Yuanshuai
Widjaja, Florensia
Zhu, Xintong
Zhou, Feige
Huang, Yongfeng
Papakonstantinou, Yannis
Ozcan, Fatma
Ma, Chenhao
Cheng, Reynold
author_facet Li, Jinyang
Li, Xiaolong
Qu, Ge
Jacobsson, Per
Qin, Bowen
Hui, Binyuan
Si, Shuzheng
Huo, Nan
Xu, Xiaohan
Zhang, Yue
Tang, Ziwei
Li, Yuanshuai
Widjaja, Florensia
Zhu, Xintong
Zhou, Feige
Huang, Yongfeng
Papakonstantinou, Yannis
Ozcan, Fatma
Ma, Chenhao
Cheng, Reynold
contents Resolution of complex SQL issues persists as a significant bottleneck in real-world database applications. Current Large Language Models (LLMs), while adept at text-to-SQL translation, have not been rigorously evaluated on the more challenging task of debugging SQL issues. To address this gap, we introduce BIRD-CRITIC, a new SQL issue debugging benchmark comprising 530 PostgreSQL tasks (BIRD-CRITIC-PG) and 570 multi-dialect tasks (BIRD-CRITIC-Multi), distilled from authentic user issues and replayed within new environments to facilitate rigorous evaluation. Baseline evaluations underscore the task's complexity, with the leading reasoning model O3-Mini achieving only 38.87% success rate on BIRD-CRITIC-PG and 33.33% on BIRD-CRITIC-Multi. Meanwhile, advancing open-source models for database tasks is crucial for empowering local development while safeguarding data privacy. Therefore, we present Six-Gym (Sql-fIX-Gym), a training environment for elevating open-source model capabilities for SQL issue debugging. This environment leverages SQL-Rewind strategy, which automatically generates executable issue-solution datasets by reverse-engineering issues from verified SQLs. However, popular trajectory-based fine-tuning methods do not explore substantial supervisory signals. We further propose f-Plan Boosting, which extracts high-level debugging plans from SQL solutions, enabling teacher LLMs to produce 73.7% more successful trajectories for training. We integrate these components into an open-source agent, Bird-Fixer. Based on Qwen-2.5-Coder-14B, Bird-Fixer achieves 38.11% success rate on BIRD-CRITIC-PG and 29.65% on BIRD-CRITIC-Multi, surpassing leading proprietary models such as Claude-3.7-Sonnet and GPT-4.1, marking a significant step toward democratizing sophisticated SQL-debugging capabilities. The leaderboard and source code are available: https://bird-critic.github.io/
format Preprint
id arxiv_https___arxiv_org_abs_2506_18951
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SWE-SQL: Illuminating LLM Pathways to Solve User SQL Issues in Real-World Applications
Li, Jinyang
Li, Xiaolong
Qu, Ge
Jacobsson, Per
Qin, Bowen
Hui, Binyuan
Si, Shuzheng
Huo, Nan
Xu, Xiaohan
Zhang, Yue
Tang, Ziwei
Li, Yuanshuai
Widjaja, Florensia
Zhu, Xintong
Zhou, Feige
Huang, Yongfeng
Papakonstantinou, Yannis
Ozcan, Fatma
Ma, Chenhao
Cheng, Reynold
Databases
Artificial Intelligence
Resolution of complex SQL issues persists as a significant bottleneck in real-world database applications. Current Large Language Models (LLMs), while adept at text-to-SQL translation, have not been rigorously evaluated on the more challenging task of debugging SQL issues. To address this gap, we introduce BIRD-CRITIC, a new SQL issue debugging benchmark comprising 530 PostgreSQL tasks (BIRD-CRITIC-PG) and 570 multi-dialect tasks (BIRD-CRITIC-Multi), distilled from authentic user issues and replayed within new environments to facilitate rigorous evaluation. Baseline evaluations underscore the task's complexity, with the leading reasoning model O3-Mini achieving only 38.87% success rate on BIRD-CRITIC-PG and 33.33% on BIRD-CRITIC-Multi. Meanwhile, advancing open-source models for database tasks is crucial for empowering local development while safeguarding data privacy. Therefore, we present Six-Gym (Sql-fIX-Gym), a training environment for elevating open-source model capabilities for SQL issue debugging. This environment leverages SQL-Rewind strategy, which automatically generates executable issue-solution datasets by reverse-engineering issues from verified SQLs. However, popular trajectory-based fine-tuning methods do not explore substantial supervisory signals. We further propose f-Plan Boosting, which extracts high-level debugging plans from SQL solutions, enabling teacher LLMs to produce 73.7% more successful trajectories for training. We integrate these components into an open-source agent, Bird-Fixer. Based on Qwen-2.5-Coder-14B, Bird-Fixer achieves 38.11% success rate on BIRD-CRITIC-PG and 29.65% on BIRD-CRITIC-Multi, surpassing leading proprietary models such as Claude-3.7-Sonnet and GPT-4.1, marking a significant step toward democratizing sophisticated SQL-debugging capabilities. The leaderboard and source code are available: https://bird-critic.github.io/
title SWE-SQL: Illuminating LLM Pathways to Solve User SQL Issues in Real-World Applications
topic Databases
Artificial Intelligence
url https://arxiv.org/abs/2506.18951