Saved in:
Bibliographic Details
Main Authors: Kim, Heegyu, Jeon, Taeyang, Choi, Seunghwan, Choi, Seungtaek, Cho, Hyunsouk
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.19014
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909367410884608
author Kim, Heegyu
Jeon, Taeyang
Choi, Seunghwan
Choi, Seungtaek
Cho, Hyunsouk
author_facet Kim, Heegyu
Jeon, Taeyang
Choi, Seunghwan
Choi, Seungtaek
Cho, Hyunsouk
contents Text-to-SQL systems have become crucial for translating natural language into SQL queries in various industries, enabling non-technical users to perform complex data operations. The need for accurate evaluation methods has increased as these systems have grown more sophisticated. However, the Execution Accuracy (EX), the most prevalent evaluation metric, still shows many false positives and negatives. Thus, this paper introduces FLEX (False-Less EXecution), a novel approach to evaluating text-to-SQL systems using large language models (LLMs) to emulate human expert-level evaluation of SQL queries. Our metric improves agreement with human experts (from 62 to 87.04 in Cohen's kappa) with comprehensive context and sophisticated criteria. Our extensive experiments yield several key insights: (1) Models' performance increases by over 2.6 points on average, substantially affecting rankings on Spider and BIRD benchmarks; (2) The underestimation of models in EX primarily stems from annotation quality issues; and (3) Model performance on particularly challenging questions tends to be overestimated. This work contributes to a more accurate and nuanced evaluation of text-to-SQL systems, potentially reshaping our understanding of state-of-the-art performance in this field.
format Preprint
id arxiv_https___arxiv_org_abs_2409_19014
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FLEX: Expert-level False-Less EXecution Metric for Reliable Text-to-SQL Benchmark
Kim, Heegyu
Jeon, Taeyang
Choi, Seunghwan
Choi, Seungtaek
Cho, Hyunsouk
Computation and Language
Information Retrieval
Machine Learning
Text-to-SQL systems have become crucial for translating natural language into SQL queries in various industries, enabling non-technical users to perform complex data operations. The need for accurate evaluation methods has increased as these systems have grown more sophisticated. However, the Execution Accuracy (EX), the most prevalent evaluation metric, still shows many false positives and negatives. Thus, this paper introduces FLEX (False-Less EXecution), a novel approach to evaluating text-to-SQL systems using large language models (LLMs) to emulate human expert-level evaluation of SQL queries. Our metric improves agreement with human experts (from 62 to 87.04 in Cohen's kappa) with comprehensive context and sophisticated criteria. Our extensive experiments yield several key insights: (1) Models' performance increases by over 2.6 points on average, substantially affecting rankings on Spider and BIRD benchmarks; (2) The underestimation of models in EX primarily stems from annotation quality issues; and (3) Model performance on particularly challenging questions tends to be overestimated. This work contributes to a more accurate and nuanced evaluation of text-to-SQL systems, potentially reshaping our understanding of state-of-the-art performance in this field.
title FLEX: Expert-level False-Less EXecution Metric for Reliable Text-to-SQL Benchmark
topic Computation and Language
Information Retrieval
Machine Learning
url https://arxiv.org/abs/2409.19014