作者位次 #5/6ACL 2026Poster· Natural Language Processing
ROSE : An Intent-Centered Evaluation Metric for NL 2 SQL
Wenqi Pei · Shizheng Hou · Boyan Li · Chen Han · Zhichao Shi · Yuyu Luo
摘要
Execution Accuracy (EX), the widely used metric for evaluating the effectiveness of Natural Language to SQL (NL2SQL) solutions, is becoming increasingly unreliable. It is sensitive to syntactic variation, ignores that questions may admit multiple interpretations, and is easily misled by erroneous...