NL2SQL is a solved problem... Not!
Summary: Despite LLM advances, enterprise-grade NL2SQL remains unsolved due to real-world challenges: complex schemata, ambiguity, robustness, integration, and evaluation gaps. Presents comparative lessons from two production teams and a roadmap for research, benchmarks, and responsible-AI practices. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Avrilia Floratou (Microsoft)
- 2. Fotis Psallidas (Microsoft)
- 3. Fuheng Zhao (University of California Santa Barbara)
- 4. Shaleen Deep (Microsoft)
- 5. Gunther Hagleither (Waii)
- 6. Wangda Tan (Waii)
- 7. Joyce Cahoon (Microsoft)
- 8. Rana Alotaibi (Microsoft)
- 9. Jordan Henkel (Microsoft)
- 10. Abhik Singla (Microsoft)
- 11. Alex van Grootel (Microsoft)
- 12. Brandon Chow (Microsoft)
- 13. Kai Deng (Microsoft)
- 14. Katherine Lin (Microsoft)
- 15. Marcos Campos (Microsoft)
- 16. Venkatesh Emani (Microsoft)
- 17. Vivek Pandit (Microsoft)
- 18. Victor Shnayder (Microsoft)
- 19. Wenjing Wang (Microsoft)
- 20. Carlo Curino (Microsoft)
BibTeX Citation
@inproceedings{floratou_cidr24,
address = {Amsterdam, Netherlands},
series = {{CIDR} '24},
title = {{NL2SQL is a solved problem... Not!}},
booktitle = {Proceedings of the {Conference} on {Innovative} {Data} {Systems} {Research}},
author = {Floratou, Avrilia and Psallidas, Fotis and Zhao, Fuheng and Deep, Shaleen and Hagleither, Gunther and Tan, Wangda and Cahoon, Joyce and Alotaibi, Rana and Henkel, Jordan and Singla, Abhik and van Grootel, Alex and Chow, Brandon and Deng, Kai and Lin, Katherine and Campos, Marcos and Emani, Venkatesh and Pandit, Vivek and Shnayder, Victor and Wang, Wenjing and Curino, Carlo},
year = {2024}
}Incoming Citations (Sorted by Pagerank)
Showing 14 of 14 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 4 of 4 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 439 | ATHENA: An Ontology-Driven System for Natural Language Querying over Relational Data Stores | 2016 | VLDB | 0.00018251249 |
| 536 | NaLIR: An Interactive Natural Language Interface for Querying Relational Databases | 2014 | SIGMOD | 0.00016771934 |
| 1,941 | Interpretable Data-Based Explanations for Fairness Debugging | 2022 | SIGMOD | 9.3297671e-05 |
| 2,353 | Bias in OLAP Queries: Detection, Explanation, and Removal (Or Think Twice About Your AVG-Query) | 2018 | SIGMOD | 8.5902514e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 12,051 | Toward Pure Natural Language Interaction with Databases | 2020 | CIDR |
| 2 | 1,356 | CatSQL: Towards Real World Natural Language to SQL Applications | 2023 | VLDB |
| 3 | 10,766 | A Comparative Evaluation of Schema Subsetting for LLM-based NL-to-SQL over Large-Schema Databases | 2026 | VLDB |
| 4 | 11,468 | Welding Natural Language Queries to Analytics IRs with LLMs | 2024 | CIDR |
| 5 | 7,952 | The Power of Constraints in Natural Language to SQL Translation | 2025 | VLDB |
| 6 | 3,489 | Combining Small Language Models and Large Language Models for Zero-Shot NL2SQL | 2024 | VLDB |
| 7 | 10,695 | NL2SQLBench: A Modular Benchmarking Framework for LLM-Enabled NL2SQL Solutions | 2026 | VLDB |
| 8 | 2,348 | The Dawn of Natural Language to SQL: Are We Fully Ready? | 2024 | VLDB |
| 9 | 776 | Natural language to SQL: Where are we today? | 2020 | VLDB |
| 10 | 9,210 | Natural Language to SQL: State of the Art and Open Problems | 2025 | VLDB |