Research paper

Auditing and Repairing LLM-as-Judge Failures in a Production Text-to-SQL Pipeline

Published by Association for Computational LinguisticsOct 2026PDF

Learn where an automated evaluator fails in a production text-to-SQL pipeline and how the evaluation was repaired.

Auditing and Repairing LLM-as-Judge Failures in a Production Text-to-SQL Pipeline cover

Resource access

Get the full resource

Enter your business details to read the full resource.

First nameLast nameWork emailCompanyJob titleCountry or region

Haowei Liu, Hsin-Tai Wu, Yi Fang. Auditing and Repairing LLM-as-Judge Failures in a Production Text-to-SQL Pipeline. Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track. © 2026 Association for Computational Linguistics. CC BY 4.0. 10.18653/2026.emnlp-industry.18. Original PDF, unmodified.

CC BY 4.0

Auditing and Repairing LLM-as-Judge Failures in a Production Text-to-SQL Pipeline

Related research.

ADEPT-SQL: A High-performance Text-to-SQL Application for Real-World Enterprise-Level Databases

Incremental Summarization for Customer Support via Progressive Note-Taking and Agent Feedback

LinkAlign: Scalable Schema Linking for Real-World Large-Scale Multi-Database Text-to-SQL