AVA: an Automatic eValuation Approach for Question Answering Systems
Abstract
We introduce AVA, an automatic evaluation approach for Question Answering, which given a set of questions associated with Gold Standard answers (references), can estimate system Accuracy. AVA uses Transformer-based language models to encode question, answer, and reference texts. This allows for effectively assessing answer correctness using similarity between the reference and an automatic answer, biased towards the question semantics. To design, train, and test AVA, we built multiple large training, development, and test sets on public and industrial benchmarks. Our innovative solutions achieve up to 74.7% F1 score in predicting human judgment for single answers. Additionally, AVA can be used to evaluate the overall system Accuracy with an error lower than 7% at 95% of confidence when measured on several QA systems.
BibTeX
@inproceedings{vu-moschitti-2021-ava,
title = "{AVA}: an Automatic e{V}aluation Approach for Question Answering Systems",
author = "Vu, Thuy and
Moschitti, Alessandro",
editor = "Toutanova, Kristina and
Rumshisky, Anna and
Zettlemoyer, Luke and
Hakkani-Tur, Dilek and
Beltagy, Iz and
Bethard, Steven and
Cotterell, Ryan and
Chakraborty, Tanmoy and
Zhou, Yichao",
booktitle = "Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
month = jun,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.naacl-main.412/",
doi = "10.18653/v1/2021.naacl-main.412",
pages = "5223--5233"
}