Quality and Quantity of Machine Translation References for Automatic Metrics
This work provides practical guidance for evaluators in shared tasks on cost-effective reference collection, though it is incremental in optimizing existing evaluation practices.
The study tackled the problem of optimizing reference quality and quantity for machine translation metrics, finding that higher-quality references improve metric correlations with humans, and mixing references from different vendors can enhance success while considering budget constraints.
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.