Improving Unsupervised Word-by-Word Translation with Language Model and Denoising Autoencoder
This work addresses the fundamental limitations of cross-lingual embeddings in sentence translation for natural language processing applications, offering a more efficient and effective unsupervised approach.
The paper tackles the problem of improving word-by-word translation using cross-lingual embeddings by integrating a language model for context-aware search and a denoising autoencoder for reordering, resulting in a system that surpasses state-of-the-art unsupervised neural translation systems without costly iterative training.
Unsupervised learning of cross-lingual word embedding offers elegant matching of words across languages, but has fundamental limitations in translating sentences. In this paper, we propose simple yet effective methods to improve word-by-word translation of cross-lingual embeddings, using only monolingual corpora but without any back-translation. We integrate a language model for context-aware search, and use a novel denoising autoencoder to handle reordering. Our system surpasses state-of-the-art unsupervised neural translation systems without costly iterative training. We also analyze the effect of vocabulary size and denoising type on the translation performance, which provides better understanding of learning the cross-lingual word embedding and its usage in translation.