Unsupervised Neural Dialect Translation with Commonality and Diversity Modeling
It addresses the problem of translating between dialects without parallel data for language communities in China, representing an incremental advance in unsupervised machine translation.
The paper tackled unsupervised dialect translation between Mandarin and Cantonese by modeling commonality and diversity, achieving over 12 BLEU score improvements compared to rule-based and conventional unsupervised methods.
As a special machine translation task, dialect translation has two main characteristics: 1) lack of parallel training corpus; and 2) possessing similar grammar between two sides of the translation. In this paper, we investigate how to exploit the commonality and diversity between dialects thus to build unsupervised translation models merely accessing to monolingual data. Specifically, we leverage pivot-private embedding, layer coordination, as well as parameter sharing to sufficiently model commonality and diversity among source and target, ranging from lexical, through syntactic, to semantic levels. In order to examine the effectiveness of the proposed models, we collect 20 million monolingual corpus for each of Mandarin and Cantonese, which are official language and the most widely used dialect in China. Experimental results reveal that our methods outperform rule-based simplified and traditional Chinese conversion and conventional unsupervised translation models over 12 BLEU scores.