Guardians of the Machine Translation Meta-Evaluation: Sentinel Metrics Fall In!

TL;DR

This paper introduces sentinel metrics to critically evaluate the meta-evaluation process of machine translation metrics, revealing biases towards certain metric types and concerns about reliance on spurious correlations.

Contribution

It proposes sentinel metrics as a novel tool to scrutinize and improve the robustness and fairness of MT metric meta-evaluation frameworks.

Findings

Current meta-evaluation favors metrics trained to mimic human judgments

Continuous metrics are disproportionately ranked highly

Potential biases and reliance on spurious correlations are identified

Abstract

Annually, at the Conference of Machine Translation (WMT), the Metrics Shared Task organizers conduct the meta-evaluation of Machine Translation (MT) metrics, ranking them according to their correlation with human judgments. Their results guide researchers toward enhancing the next generation of metrics and MT systems. With the recent introduction of neural metrics, the field has witnessed notable advancements. Nevertheless, the inherent opacity of these metrics has posed substantial challenges to the meta-evaluation process. This work highlights two issues with the meta-evaluation framework currently employed in WMT, and assesses their impact on the metrics rankings. To do this, we introduce the concept of sentinel metrics, which are designed explicitly to scrutinize the meta-evaluation process's accuracy, robustness, and fairness. By employing sentinel metrics, we aim to validate our findings, and shed light on and monitor the potential biases or inconsistencies in the rankings. We discover that the present meta-evaluation framework favors two categories of metrics: i) those explicitly trained to mimic human quality assessments, and ii) continuous metrics. Finally, we raise concerns regarding the evaluation capabilities of state-of-the-art metrics, emphasizing that they might be basing their assessments on spurious correlations found in their training data.