# TrustSQL: Benchmarking Text-to-SQL Reliability with Penalty-Based Scoring

arXiv: https://arxiv.org/abs/2403.15879 · HTML: https://mlsanity.com/paper/2403.15879/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 30 references that are in the map.
- First posted on arXiv 23 Mar 2024.

## Builds on (most cited first, 24 of 24)

- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks](https://mlsanity.com/paper/1807.03888/) (2018)
- [Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task](https://mlsanity.com/paper/1809.08887/) (2018)
- [Teaching Large Language Models to Self-Debug](https://mlsanity.com/paper/2304.05128/) (2023)
- [Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs](https://mlsanity.com/paper/2305.03111/) (2023)
- [Tabular Data: Deep Learning is Not All You Need](https://mlsanity.com/paper/2106.03253/) (2021)
- [DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correction](https://mlsanity.com/paper/2304.11015/) (2023)
- [Unsolved Problems in ML Safety](https://mlsanity.com/paper/2109.13916/) (2021)
- [A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection](https://mlsanity.com/paper/2106.09022/) (2021)
- [Out-of-Distribution Detection and Selective Generation for Conditional Language Models](https://mlsanity.com/paper/2209.15558/) (2022)
- [On Hallucination and Predictive Uncertainty in Conditional Language Generation](https://mlsanity.com/paper/2103.15025/) (2021)
- [Improving Text-to-SQL Evaluation Methodology](https://mlsanity.com/paper/1806.09029/) (2018)
- [KaggleDBQA: Realistic Evaluation of Text-to-SQL Parsers](https://mlsanity.com/paper/2106.11455/) (2021)
- [Reliable Visual Question Answering: Abstain Rather Than Answer Incorrectly](https://mlsanity.com/paper/2204.13631/) (2022)
- [Out-of-domain Detection for Natural Language Understanding in Dialog Systems](https://mlsanity.com/paper/1909.03862/) (2019)
- [How to Prompt LLMs for Text-to-SQL: A Study in Zero-shot, Single-domain, and Cross-domain Settings](https://mlsanity.com/paper/2305.11853/) (2023)
- [EHRSQL: A Practical Text-to-SQL Benchmark for Electronic Health Records](https://mlsanity.com/paper/2301.07695/) (2023)
- [Text-to-SQL in the Wild: A Naturally-Occurring Dataset Based on Stack Exchange Data](https://mlsanity.com/paper/2106.05006/) (2021)
- [Calibrated Interpretation: Confidence Estimation in Semantic Parsing](https://mlsanity.com/paper/2211.07443/) (2022)
- [ConditionalQA: A Complex Reading Comprehension Dataset with Conditional Answers](https://mlsanity.com/paper/2110.06884/) (2021)
- [Towards Robustness of Text-to-SQL Models Against Natural and Realistic Adversarial Table Perturbation](https://mlsanity.com/paper/2212.09994/) (2022)
- [Did You Ask a Good Question? A Cross-Domain Question Intention Classification Benchmark for Text-to-SQL](https://mlsanity.com/paper/2010.12634/) (2020)
- [OodGAN: Generative Adversarial Network for Out-of-Domain Data Generation](https://mlsanity.com/paper/2104.02484/) (2021)
- [Uncertainty-Aware Text-to-Program for Question Answering on Structured Electronic Health Records](https://mlsanity.com/paper/2203.06918/) (2022)

## Built on by (most cited first, 17 of 17)

- [Overview of the EHRSQL 2024 Shared Task on Reliable Text-to-SQL Modeling on Electronic Health Records](https://mlsanity.com/paper/2405.06673/) (2024)
- [Confidence Estimation for Error Detection in Text-to-SQL Systems](https://mlsanity.com/paper/2501.09527/) (2025)
- [Tabular Data Understanding with LLMs: A Survey of Recent Advances and Challenges](https://mlsanity.com/paper/2508.00217/) (2025)
- [ProbGate at EHRSQL 2024: Enhancing SQL Query Generation Accuracy through Probabilistic Threshold Filtering and Error Handling](https://mlsanity.com/paper/2404.16659/) (2024)
- [QDA-SQL: Questions Enhanced Dialogue Augmentation for Multi-Turn Text-to-SQL](https://mlsanity.com/paper/2406.10593/) (2024)
- [SAFENLIDB: A Privacy-Preserving Safety Alignment Framework for LLM-based Natural Language Database Interfaces](https://mlsanity.com/paper/2511.06778/) (2025)
- [M3: Conversational LLMs Simplify Secure Clinical Data Access, Understanding, and Analysis](https://mlsanity.com/paper/2507.01053/) (2025)
- [Business Truth, not SQL Accuracy: A Rule-Gated 7B Analytics Agent Outperforms a Direct-Prompted 32B Baseline](https://mlsanity.com/paper/2608.09254/) (2026)
- [Never the Number: Structural Abstention for AI Systems Whose Answers Are Consumed as Fact](https://mlsanity.com/paper/2608.13926/) (2026)
- [The Stochastic Shift: A New Evaluation Paradigm for Text-to-SQL with AI Operators](https://mlsanity.com/paper/2609.21133/) (2026)
- [SCARE: A Benchmark for SQL Correction and Question Answerability Classification for Reliable EHR Question Answering](https://mlsanity.com/paper/2511.17559/) (2025)
- [ReliableTableQA:How Much Supervision Does Reliability Annotation Need?](https://mlsanity.com/paper/2607.20537/) (2026)
- [Bounded Semantic Planning and Deterministic Compilation for Reliable Enterprise Text-to-SQL](https://mlsanity.com/paper/2608.16663/) (2026)
- [TrustDABench: Benchmarking Reliability and Robustness of LLMs for Structured Data Analysis](https://mlsanity.com/paper/2608.24145/) (2026)
- [Certified Against Which Oracle? Execution Labels Set the Reported Risk of Conformal Abstention for Text-to-SQL](https://mlsanity.com/paper/2609.25938/) (2026)
- [Answering clinicians' questions over trial evidence tables with verifiable, feedback-driven language models](https://mlsanity.com/paper/2610.02576/) (2026)
- [Trajectory-Derived Confidence for Reliable, Resource-Aware Clinical Text-to-SQL Agents](https://mlsanity.com/paper/2610.04156/) (2026)

## Abstract (arXiv)

Text-to-SQL enables users to interact with databases using natural language, simplifying the retrieval and synthesis of information. Despite the remarkable success of large language models (LLMs) in translating natural language questions into SQL queries, widespread deployment remains limited due to two primary challenges. First, the effective use of text-to-SQL models depends on users' understanding of the model's capabilities-the scope of questions the model can correctly answer. Second, the absence of abstention mechanisms can lead to incorrect SQL generation going unnoticed, thereby undermining trust in the model's output. To enable wider deployment, it is crucial to address these challenges in model design and enhance model evaluation to build trust in the model's output. To this end, we introduce TrustSQL, a novel comprehensive benchmark designed to evaluate text-to-SQL reliability-defined as a model's ability to correctly handle any type of input question by generating correct SQL queries for feasible questions and abstaining from generating infeasible ones (e.g., due to schema incompatibility or functionalities beyond SQL). We evaluate existing methods using a novel penalty-based scoring metric with two modeling approaches: (1) pipeline-based methods combining SQL generators with infeasible question detectors and SQL error detectors for abstention; and (2) unified methods using a single model for the entire task. Our experimental results reveal that achieving high scores under severe penalties requires significant effort and provide a new perspective on developing text-to-SQL models for safer deployment. TrustSQL is available at https://github.com/glee4810/TrustSQL.