Quality and Quantity of Machine Translation References for Automatic Metrics

Vilém Zouhar, Ondřej Bojar


Abstract
Automatic machine translation metrics typically rely on human translations to determine the quality of system translations. Common wisdom in the field dictates that the human references should be of very high quality. However, there are no cost-benefit analyses that could be used to guide practitioners who plan to collect references for machine translation evaluation. We find that higher-quality references lead to better metric correlations with humans at the segment-level. Having up to 7 references per segment and taking their average (or maximum) helps all metrics. Interestingly, the references from vendors of different qualities can be mixed together and improve metric success. Higher quality references, however, cost more to create and we frame this as an optimization problem: given a specific budget, what references should be collected to maximize metric success. These findings can be used by evaluators of shared tasks when references need to be created under a certain budget.
Anthology ID:
2024.humeval-1.1
Volume:
Proceedings of the Fourth Workshop on Human Evaluation of NLP Systems (HumEval) @ LREC-COLING 2024
Month:
May
Year:
2024
Address:
Torino, Italia
Editors:
Simone Balloccu, Anya Belz, Rudali Huidrom, Ehud Reiter, Joao Sedoc, Craig Thomson
Venues:
HumEval | WS
SIG:
Publisher:
ELRA and ICCL
Note:
Pages:
1–11
Language:
URL:
https://aclanthology.org/2024.humeval-1.1
DOI:
Bibkey:
Cite (ACL):
Vilém Zouhar and Ondřej Bojar. 2024. Quality and Quantity of Machine Translation References for Automatic Metrics. In Proceedings of the Fourth Workshop on Human Evaluation of NLP Systems (HumEval) @ LREC-COLING 2024, pages 1–11, Torino, Italia. ELRA and ICCL.
Cite (Informal):
Quality and Quantity of Machine Translation References for Automatic Metrics (Zouhar & Bojar, HumEval-WS 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.humeval-1.1.pdf