Sphinteract: Resolving Ambiguities in NL2SQL Through User Interaction
Summary: Sphinteract: an interactive NL2SQL framework that resolves user-question ambiguities by eliciting minimal feedback via a Summarize–Review–Ask (SRA) loop to prompt LLMs for targeted clarifications. Proposes three feedback-integration methods and reports up to 42% accuracy gains on KaggleDBQA and BIRD, showing interactive disambiguation outperforms pure prompt/tuning approaches. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Fuheng Zhao (University of California Santa Barbara)
- 2. Shaleen Deep (Microsoft)
- 3. Fotis Psallidas (Microsoft)
- 4. Avrilia Floratou (Microsoft)
- 5. Divyakant Agrawal (University of California Santa Barbara)
- 6. Amr El Abbadi (University of California Santa Barbara)
BibTeX Citation
@article{zhao_vldb25,
title = {{Sphinteract: Resolving Ambiguities in NL2SQL Through User Interaction}},
author = {Zhao, Fuheng and Deep, Shaleen and Psallidas, Fotis and Floratou, Avrilia and Agrawal, Divyakant and Abbadi, Amr El},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {4},
pages = {1145--1158},
doi = {10.14778/3717755.3717772},
url = {https://doi.org/10.14778/3717755.3717772},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 2 of 2 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,506 | This is Going to Sound Crazy, But What If We Used Large Language Models to Boost Automatic Database Tuning Algorithms By Leveraging Prior History? We Will Find Better Configurations More Quickly Than Retraining From Scratch! | 2026 | SIGMOD | 5.093636e-05 |
| 10,510 | NL2SQLBench: A Modular Benchmarking Framework for LLM-Enabled NL2SQL Solutions | 2026 | VLDB | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 12 of 12 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 11,120 | Welding Natural Language Queries to Analytics IRs with LLMs | 2024 | CIDR |
| 2 | 2,048 | Duoquest: A Dual-Specification System for Expressive SQL Queries | 2020 | SIGMOD |
| 3 | 10,732 | RTS+: Reliable Text to SQL | 2025 | SIGMOD |
| 4 | 8,987 | Constructing Expressive Relational Queries with Dual-Specification Synthesis | 2020 | CIDR |
| 5 | 3,787 | Combining Small Language Models and Large Language Models for Zero-Shot NL2SQL | 2024 | VLDB |
| 6 | 10,510 | NL2SQLBench: A Modular Benchmarking Framework for LLM-Enabled NL2SQL Solutions | 2026 | VLDB |
| 7 | 9,870 | Natural Language to SQL: State of the Art and Open Problems | 2025 | VLDB |
| 8 | 2,852 | The Dawn of Natural Language to SQL: Are We Fully Ready? | 2024 | VLDB |
| 9 | 6,936 | Reliable Text-to-SQL with Adaptive Abstention | 2025 | SIGMOD |
| 10 | 2,602 | NL2SQL is a solved problem... Not! | 2024 | CIDR |