Adapting to Non-Centered Languages for Zero-shot Multilingual Translation

COLING 2022  ·  Zhi Qu, Taro Watanabe ·

Multilingual neural machine translation can translate unseen language pairs during training, i.e. zero-shot translation. However, the zero-shot translation is always unstable. Although prior works attributed the instability to the domination of central language, e.g. English, we supplement this viewpoint with the strict dependence of non-centered languages. In this work, we propose a simple, lightweight yet effective language-specific modeling method by adapting to non-centered languages and combining the shared information and the language-specific information to counteract the instability of zero-shot translation. Experiments with Transformer on IWSLT17, Europarl, TED talks, and OPUS-100 datasets show that our method not only performs better than strong baselines in centered data conditions but also can easily fit non-centered data conditions. By further investigating the layer attribution, we show that our proposed method can disentangle the coupled representation in the correct direction.

PDF Abstract COLING 2022 PDF COLING 2022 Abstract


Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.