An Empirical Study on the Robustness of Massively Multilingual Neural Machine Translation
This work addresses translation robustness for a specific language pair, providing an incremental contribution with a new benchmark dataset.
The paper investigated the robustness of massively multilingual neural machine translation for Indonesian-Chinese translation under naturally occurring noise, creating a benchmark dataset and evaluating four NLLB-200 models to reveal correlations between error types, noise types, model sizes, and evaluation metrics.
Massively multilingual neural machine translation (MMNMT) has been proven to enhance the translation quality of low-resource languages. In this paper, we empirically investigate the translation robustness of Indonesian-Chinese translation in the face of various naturally occurring noise. To assess this, we create a robustness evaluation benchmark dataset for Indonesian-Chinese translation. This dataset is automatically translated into Chinese using four NLLB-200 models of different sizes. We conduct both automatic and human evaluations. Our in-depth analysis reveal the correlations between translation error types and the types of noise present, how these correlations change across different model sizes, and the relationships between automatic evaluation indicators and human evaluation indicators. The dataset is publicly available at https://github.com/tjunlp-lab/ID-ZH-MTRobustEval.