Saved in:
Bibliographic Details
Main Authors: Park, ChaeHun, Choi, Minseok, Lee, Dohyun, Choo, Jaegul
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.01015
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910532907302912
author Park, ChaeHun
Choi, Minseok
Lee, Dohyun
Choo, Jaegul
author_facet Park, ChaeHun
Choi, Minseok
Lee, Dohyun
Choo, Jaegul
contents Building a reliable and automated evaluation metric is a necessary but challenging problem for open-domain dialogue systems. Recent studies proposed evaluation metrics that assess generated responses by considering their relevance to previous dialogue histories. Although effective, these metrics evaluate individual responses directly rather than considering their relative quality compared to other responses. To handle this, we propose PairEval, a novel dialogue evaluation metric for assessing responses by comparing their quality against responses in different conversations. PairEval is built on top of open-sourced and moderate-size language models, and we make them specialized in pairwise comparison between dialogue responses. Extensive experiments on multiple benchmarks demonstrate that our metric exhibits a higher correlation with human judgments than baseline metrics. We also find that the proposed comparative metric is more robust in detecting common failures from open-domain dialogue systems, including repetition and speaker insensitivity.
format Preprint
id arxiv_https___arxiv_org_abs_2404_01015
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle PairEval: Open-domain Dialogue Evaluation with Pairwise Comparison
Park, ChaeHun
Choi, Minseok
Lee, Dohyun
Choo, Jaegul
Computation and Language
Building a reliable and automated evaluation metric is a necessary but challenging problem for open-domain dialogue systems. Recent studies proposed evaluation metrics that assess generated responses by considering their relevance to previous dialogue histories. Although effective, these metrics evaluate individual responses directly rather than considering their relative quality compared to other responses. To handle this, we propose PairEval, a novel dialogue evaluation metric for assessing responses by comparing their quality against responses in different conversations. PairEval is built on top of open-sourced and moderate-size language models, and we make them specialized in pairwise comparison between dialogue responses. Extensive experiments on multiple benchmarks demonstrate that our metric exhibits a higher correlation with human judgments than baseline metrics. We also find that the proposed comparative metric is more robust in detecting common failures from open-domain dialogue systems, including repetition and speaker insensitivity.
title PairEval: Open-domain Dialogue Evaluation with Pairwise Comparison
topic Computation and Language
url https://arxiv.org/abs/2404.01015