An Empirical study of Unsupervised Neural Machine Translation: analyzing NMT output, model's behavior and sentences' contribution
This provides empirical analysis of UNMT's behavior and limitations for researchers working on low-resource translation, though it appears incremental in expanding existing analysis methods to the UNMT paradigm.
This paper analyzed unsupervised neural machine translation (UNMT) across three diverse languages (French, Gujarati, Kazakh) by comparing output quality, word order, and semantic similarity to supervised methods in high- and low-resource setups, while using Layer-wise Relevance Propagation to examine sentence contributions.
Unsupervised Neural Machine Translation (UNMT) focuses on improving NMT results under the assumption there is no human translated parallel data, yet little work has been done so far in highlighting its advantages compared to supervised methods and analyzing its output in aspects other than translation accuracy. We focus on three very diverse languages, French, Gujarati, and Kazakh, and train bilingual NMT models, to and from English, with various levels of supervision, in high- and low- resource setups, measure quality of the NMT output and compare the generated sequences' word order and semantic similarity to source and reference sentences. We also use Layer-wise Relevance Propagation to evaluate the source and target sentences' contribution to the result, expanding the findings of previous works to the UNMT paradigm.