SQUARE: Automatic Question Answering Evaluation using Multiple Positive and Negative References

TL;DR

SQuArE is a new QA evaluation metric that leverages multiple correct and incorrect reference answers, improving correlation with human judgment for both extractive and generative QA systems.

Contribution

It introduces a novel evaluation method using multiple references, addressing limitations of previous single-reference metrics in QA system assessment.

Findings

Outperforms previous metrics in correlation with human annotations

Effective for both extractive and generative QA tasks

Validated across multiple datasets and settings

Abstract

Evaluation of QA systems is very challenging and expensive, with the most reliable approach being human annotations of correctness of answers for questions. Recent works (AVA, BEM) have shown that transformer LM encoder based similarity metrics transfer well for QA evaluation, but they are limited by the usage of a single correct reference answer. We propose a new evaluation metric: SQuArE (Sentence-level QUestion AnsweRing Evaluation), using multiple reference answers (combining multiple correct and incorrect references) for sentence-form QA. We evaluate SQuArE on both sentence-level extractive (Answer Selection) and generative (GenQA) QA systems, across multiple academic and industrial datasets, and show that it outperforms previous baselines and obtains the highest correlation with human annotations.