@inproceedings{lakew-etal-2021-zero,
    title = "Zero-Shot Neural Machine Translation with Self-Learning Cycle",
    author = "Lakew, Surafel M.  and
      Negri, Matteo  and
      Turchi, Marco",
    editor = "Ortega, John  and
      Ojha, Atul Kr.  and
      Kann, Katharina  and
      Liu, Chao-Hong",
    booktitle = "Proceedings of the 4th Workshop on Technologies for MT of Low Resource Languages (LoResMT2021)",
    month = aug,
    year = "2021",
    address = "Virtual",
    publisher = "Association for Machine Translation in the Americas",
    url = "https://preview.aclanthology.org/add-emnlp-2024-awards/2021.mtsummit-loresmt.10/",
    pages = "96--113",
    abstract = "Neural Machine Translation (NMT) approaches employing monolingual data are showing steady improvements in resource-rich conditions. However, evaluations using real-world lowresource languages still result in unsatisfactory performance. This work proposes a novel zeroshot NMT modeling approach that learns without the now-standard assumption of a pivot language sharing parallel data with the zero-shot source and target languages. Our approach is based on three stages: initialization from any pre-trained NMT model observing at least the target language, augmentation of source sides leveraging target monolingual data, and learning to optimize the initial model to the zero-shot pair, where the latter two constitute a selflearning cycle. Empirical findings involving four diverse (in terms of a language family, script and relatedness) zero-shot pairs show the effectiveness of our approach with up to +5.93 BLEU improvement against a supervised bilingual baseline. Compared to unsupervised NMT, consistent improvements are observed even in a domain-mismatch setting, attesting to the usability of our method."
}