Document detail
ID
oai:arXiv.org:2407.12832
Topic
Computer Science - Computation and...Author
Cavalin, Paulo Domingues, Pedro Henrique Pinhanez, ClaudioCategory
Computer Science
Year
2024
listing date
7/24/2024
Abstract
In this paper we show that corpus-level aggregation hinders considerably the capability of lexical metrics to accurately evaluate machine translation (MT) systems.
With empirical experiments we demonstrate that averaging individual segment-level scores can make metrics such as BLEU and chrF correlate much stronger with human judgements and make them behave considerably more similar to neural metrics such as COMET and BLEURT.
We show that this difference exists because corpus- and segment-level aggregation differs considerably owing to the classical average of ratio versus ratio of averages Mathematical problem.
Moreover, as we also show, such difference affects considerably the statistical robustness of corpus-level aggregation.
Considering that neural metrics currently only cover a small set of sufficiently-resourced languages, the results in this paper can help make the evaluation of MT systems for low-resource languages more trustworthy.
Cavalin, Paulo,Domingues, Pedro Henrique,Pinhanez, Claudio, 2024, Sentence-level Aggregation of Lexical Metrics Correlate Stronger with Human Judgements than Corpus-level Aggregation
Leveraging Bi-Focal Perspectives and Granular Feature Integration for Accurate Reliable Early Alzheimer's Detection
ad computer information science model
