TY - GEN
T1 - DATScore
T2 - 17th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2023 - Findings of EACL 2023
AU - Eddine, Moussa Kamal
AU - Shang, Guokan
AU - Vazirgiannis, Michalis
N1 - Publisher Copyright:
© 2023 Association for Computational Linguistics.
PY - 2023/1/1
Y1 - 2023/1/1
N2 - The rapid development of large pretrained language models has revolutionized not only the field of Natural Language Generation (NLG) but also its evaluation. Inspired by the recent work of BARTScore: a metric leveraging the BART language model to evaluate the quality of generated text from various aspects, we introduce DATScore. DATScore uses data augmentation techniques to improve the evaluation of machine translation. Our main finding is that introducing data augmented translations of the source and reference texts is greatly helpful in evaluating the quality of the generated translation. We also propose two novel score averaging and term weighting strategies to improve the original score computing process of BARTScore. Experimental results on WMT show that DATScore correlates better with human meta-evaluations than the other recent state-of-the-art metrics, especially for low-resource languages. Ablation studies demonstrate the value added by our new scoring strategies.
AB - The rapid development of large pretrained language models has revolutionized not only the field of Natural Language Generation (NLG) but also its evaluation. Inspired by the recent work of BARTScore: a metric leveraging the BART language model to evaluate the quality of generated text from various aspects, we introduce DATScore. DATScore uses data augmentation techniques to improve the evaluation of machine translation. Our main finding is that introducing data augmented translations of the source and reference texts is greatly helpful in evaluating the quality of the generated translation. We also propose two novel score averaging and term weighting strategies to improve the original score computing process of BARTScore. Experimental results on WMT show that DATScore correlates better with human meta-evaluations than the other recent state-of-the-art metrics, especially for low-resource languages. Ablation studies demonstrate the value added by our new scoring strategies.
U2 - 10.18653/v1/2023.findings-eacl.69
DO - 10.18653/v1/2023.findings-eacl.69
M3 - Conference contribution
AN - SCOPUS:85159860710
T3 - EACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2023
SP - 912
EP - 922
BT - EACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2023
PB - Association for Computational Linguistics (ACL)
Y2 - 2 May 2023 through 6 May 2023
ER -