MT-TeQL: Evaluating and Augmenting Neural NLIDB on Real-world Linguistic and Schema Variations
Summary: MT-TeQL applies metamorphic testing to NLIDBs, with semantics-preserving utterance/schema transformations to expose robustness gaps. Across 9 NLIDBs on 62,430 inputs, 15,433 defects; MT-TeQL variants reduce errors by 46.5% without lowering accuracy. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Pingchuan Ma
- 2. Shuai Wang
Incoming Citations (Sorted by Pagerank)
Showing 7 of 7 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,731 | CatSQL: Towards Real World Natural Language to SQL Applications | 2023 | VLDB | 0.00010723825 |
| 2,435 | ScienceBenchmark: A Complex Real-World Benchmark for Evaluating Natural Language to SQL Systems | 2024 | VLDB | 8.8218963e-05 |
| 4,505 | Testing Graph Database Systems via Graph-Aware Metamorphic Relations | 2024 | VLDB | 6.1290872e-05 |
| 5,448 | SNAILS: Schema Naming Assessments for Improved LLM-Based SQL Inference | 2025 | SIGMOD | 5.4980173e-05 |
| 10,701 | Evoschema: Towards Text-To-Sql Robustness Against Schema Evolution | 2025 | VLDB | 4.1905499e-05 |
| 10,841 | Natural Language to SQL: State of the Art and Open Problems | 2025 | VLDB | 4.1905499e-05 |
| 10,848 | Panel on Neural Relational Data: Tabular Foundation Models, LLMs... or both? | 2025 | VLDB | 4.1905499e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 5 of 5 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 205 | Constructing an Interactive Natural Language Interface for Relational Databases | 2015 | VLDB | 0.00034651121 |
| 535 | ATHENA: An Ontology-Driven System for Natural Language Querying over Relational Data Stores | 2016 | VLDB | 0.00020718836 |
| 973 | Natural language to SQL: Where are we today? | 2020 | VLDB | 0.0001488435 |
| 1,047 | Functional Dependency Discovery: An Experimental Evaluation of Seven Algorithms | 2015 | VLDB | 0.00014442219 |
| 2,325 | DBPal: A Fully Pluggable NL2SQL Training Pipeline | 2020 | SIGMOD | 9.0277894e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| Overall Rank | Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,841 | Natural Language to SQL: State of the Art and Open Problems | 2025 | VLDB | 4.1905499e-05 |
| 3,189 | Text2SQL is Not Enough: Unifying AI and Databases with TAG | 2025 | CIDR | 7.4140094e-05 |
| 3,666 | The Dawn of Natural Language to SQL: Are We Fully Ready? | 2024 | VLDB | 6.8606092e-05 |
| 11,555 | Toward Pure Natural Language Interaction with Databases | 2020 | CIDR | 4.1905499e-05 |
| 10,701 | Evoschema: Towards Text-To-Sql Robustness Against Schema Evolution | 2025 | VLDB | 4.1905499e-05 |
| 7,351 | Reliable Text-to-SQL with Adaptive Abstention | 2025 | SIGMOD | 4.7484027e-05 |
| 2,025 | From Natural Language Processing to Neural Databases | 2021 | VLDB | 9.7477788e-05 |
| 10,221 | NL2SQLBench: A Modular Benchmarking Framework for LLM-Enabled NL2SQL Solutions | 2026 | VLDB | 4.1905499e-05 |
| 1,731 | CatSQL: Towards Real World Natural Language to SQL Applications | 2023 | VLDB | 0.00010723825 |
| 973 | Natural language to SQL: Where are we today? | 2020 | VLDB | 0.0001488435 |