# Certified Against Which Oracle? Execution Labels Set the Reported Risk of Conformal Abstention for Text-to-SQL

arXiv: https://arxiv.org/abs/2609.25938 · HTML: https://mlsanity.com/paper/2609.25938/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 48 of its 56 references that are in the map.
- First posted on arXiv 22 Sep 2026.

## Builds on (most cited first, 30 of 48)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [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)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation](https://mlsanity.com/paper/2302.09664/) (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)
- [Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarks](https://mlsanity.com/paper/2103.14749/) (2021)
- [Conformal Risk Control](https://mlsanity.com/paper/2208.02814/) (2022)
- [Generating with Confidence: Uncertainty Quantification for Black-box Large Language Models](https://mlsanity.com/paper/2305.19187/) (2023)
- [Learn then Test: Calibrating Predictive Algorithms to Achieve Risk Control](https://mlsanity.com/paper/2110.01052/) (2021)
- [Conformal Language Modeling](https://mlsanity.com/paper/2306.10193/) (2023)
- [LM-Polygraph: Uncertainty Estimation for Language Models](https://mlsanity.com/paper/2311.07383/) (2023)
- [Language Models with Conformal Factuality Guarantees](https://mlsanity.com/paper/2402.10978/) (2024)
- [Semantic Evaluation for Text-to-SQL with Distilled Test Suites](https://mlsanity.com/paper/2010.02840/) (2020)
- [Structure-Grounded Pretraining for Text-to-SQL](https://mlsanity.com/paper/2010.12773/) (2020)
- [OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scale](https://mlsanity.com/paper/2503.02240/) (2025)
- [XiYan-SQL: A Novel Multi-Generator Framework For Text-to-SQL](https://mlsanity.com/paper/2507.04701/) (2025)
- [Understanding the Effects of Noise in Text-to-SQL: An Examination of the BIRD-Bench Benchmark](https://mlsanity.com/paper/2402.12243/) (2024)
- [Calibrated Interpretation: Confidence Estimation in Semantic Parsing](https://mlsanity.com/paper/2211.07443/) (2022)
- [Evaluating Cross-Domain Text-to-SQL Models and Benchmarks](https://mlsanity.com/paper/2310.18538/) (2023)
- [Adaptive conformal classification with noisy labels](https://mlsanity.com/paper/2309.05092/) (2023)
- [Label Noise Robustness of Conformal Prediction](https://mlsanity.com/paper/2209.14295/) (2022)
- [Reliable Text-to-SQL with Adaptive Abstention](https://mlsanity.com/paper/2501.10858/) (2025)
- [Uncertainty Quantification for Language Models: A Suite of Black-Box, White-Box, LLM Judge, and Ensemble Scorers](https://mlsanity.com/paper/2504.19254/) (2025)
- [TrustSQL: Benchmarking Text-to-SQL Reliability with Penalty-Based Scoring](https://mlsanity.com/paper/2403.15879/) (2024)
- [Universal distribution of the empirical coverage in split conformal prediction](https://mlsanity.com/paper/2303.02770/) (2023)
- [Pervasive Annotation Errors Break Text-to-SQL Benchmarks and Leaderboards](https://mlsanity.com/paper/2601.08778/) (2026)
- [Error Detection for Text-to-SQL Semantic Parsing](https://mlsanity.com/paper/2305.13683/) (2023)
- [FLEX: Expert-level False-Less EXecution Metric for Reliable Text-to-SQL Benchmark](https://mlsanity.com/paper/2409.19014/) (2024)

## Abstract (arXiv)

A conformal abstention certificate for text-to-SQL is only as truthful as the correctness labels it is calibrated on. The uncertainty pipelines that read confidence off execution consistency take those labels from the single database a benchmark ships, an oracle known to be lenient. We run a preregistered intervention on Spider-Realistic, swapping that database for the benchmark's distilled multi-instance test suite. Across four SQL-specialist checkpoints and two split schemes, the swap raises the certificate's held-out risk 2.73 to 10.23 points above the risk its own labels report. Neither oracle reports the risk experts assign. Under blinded labels from two SQL experts, a certificate calibrated at a nominal 0.10 carries 20.0 and 17.2 points of risk on two checkpoints. The stricter oracle errs in both directions: most of the answers it rejects are not judged wrong, and some of those it accepts are. An AI-assigned census of what it rejects finds a semantic error in a quarter to a third of them, depending on the population. It attributes most of the rest to underspecified questions, synthetic instances or suspected reference-query defects, a flag supported by a preregistered blinded expert audit. The oracle also decides how a confidence score is judged. Every execution-consistency score looks better under the labels of the oracle that built its clusters, in 16 of 16 combinations. Under expert labels, building such a score on suite clusters instead of shipped-database clusters raises its area under the ROC curve (AUROC) by 6.96 points on one checkpoint and 1.53 on the other. On the second, the expert interval excludes the 8.3 points the suite labels report. A certificate should be reported with both oracles, and an oracle-relative difference read as semantic risk only after the benchmark is audited. A consistency score should be evaluated under an oracle that did not build it.