TACO: A Benchmark for Open-Domain Text-to-SQL with Ambiguous and Cross-Database Queries
Summary: TACO benchmarks open-domain text-to-SQL beyond standard closed-schema settings, targeting ambiguous questions, unspecified databases, and cross-database queries. It combines 1.5K real smart-city examples with 13K synthesized open-data queries, plus a TACO-SQL baseline revealing a large gap to human SQL. (summarized by gpt-5.4-mini on Apr 12 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Chao Deng (Renmin University of China)
- 2. Ju Fan (Renmin University of China)
- 3. Yuyu Luo (Hong Kong University of Science and Technology)
- 4. Qinliang Xue (Renmin University of China)
- 5. Meihao Fan (Renmin University of China)
- 6. Yuxin Zhang (Renmin University of China)
- 7. Min Zhang (Beijing Academy of Artificial Intelligence; Renmin University of China)
- 8. Xiaofeng Jia (Beijing Academy of Artificial Intelligence)
- 9. Jing Zhang (Renmin University of China)
- 10. Xiaoyong Du (Renmin University of China)
BibTeX Citation
@article{deng_vldb26,
title = {{TACO: A Benchmark for Open-Domain Text-to-SQL with Ambiguous and Cross-Database Queries}},
author = {Deng, Chao and Fan, Ju and Luo, Yuyu and Xue, Qinliang and Fan, Meihao and Zhang, Yuxin and Zhang, Min and Jia, Xiaofeng and Zhang, Jing and Du, Xiaoyong},
journal = {PVLDB},
series = {{VLDB} '26},
volume = {19},
number = {6},
pages = {1386--1399},
doi = {10.14778/3797919.3797942},
url = {https://doi.org/10.14778/3797919.3797942},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 7 of 7 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 756 | CodeS: Towards Building Open-source Language Models for Text-to-SQL | 2024 | SIGMOD | 0.0001431656 |
| 865 | Natural language to SQL: Where are we today? | 2020 | VLDB | 0.00013521464 |
| 2,748 | Few-shot Text-to-SQL Translation using Structure and Content Prompt Learning | 2023 | SIGMOD | 8.1707811e-05 |
| 3,787 | Combining Small Language Models and Large Language Models for Zero-Shot NL2SQL | 2024 | VLDB | 7.1249098e-05 |
| 5,323 | SNAILS: Schema Naming Assessments for Improved LLM-Based SQL Inference | 2025 | SIGMOD | 6.2655413e-05 |
| 10,285 | Reward-SQL: Boosting Text-to-SQL via Stepwise Execution-Aware Reasoning and Process-Supervised Rewards | 2026 | SIGMOD | 5.093636e-05 |
| 10,931 | AutoPrep: Natural Language Question-Aware Data Preparation with a Multi-Agent Framework | 2025 | VLDB | 5.093636e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 10,530 | SQL-Exchange: Transforming SQL Queries Across Domains | 2026 | VLDB |
| 2 | 2,748 | Few-shot Text-to-SQL Translation using Structure and Content Prompt Learning | 2023 | SIGMOD |
| 3 | 10,732 | RTS+: Reliable Text to SQL | 2025 | SIGMOD |
| 4 | 756 | CodeS: Towards Building Open-source Language Models for Text-to-SQL | 2024 | SIGMOD |
| 5 | 865 | Natural language to SQL: Where are we today? | 2020 | VLDB |
| 6 | 279 | Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation | 2024 | VLDB |
| 7 | 2,395 | OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scale | 2025 | VLDB |
| 8 | 2,809 | Text2SQL is Not Enough: Unifying AI and Databases with TAG | 2025 | CIDR |
| 9 | 10,556 | OpenSQL: Data-Efficient Text-to-SQL for Open-Source LLMs via Synthesized Intermediate Supervision | 2026 | VLDB |
| 10 | 5,174 | An In-Depth Benchmarking of Text-to-SQL Systems | 2021 | SIGMOD |