@inproceedings{ai-fang-2023-fly,
    title = "On-the-fly Cross-lingual Masking for Multilingual Pre-training",
    author = "Ai, Xi  and
      Fang, Bin",
    editor = "Rogers, Anna  and
      Boyd-Graber, Jordan  and
      Okazaki, Naoaki",
    booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2023",
    address = "Toronto, Canada",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2023.acl-long.49/",
    doi = "10.18653/v1/2023.acl-long.49",
    pages = "855--876",
    abstract = "In multilingual pre-training with the objective of MLM (masked language modeling) on multiple monolingual corpora, multilingual models only learn cross-linguality implicitly from isomorphic spaces formed by overlapping different language spaces due to the lack of explicit cross-lingual forward pass. In this work, we present CLPM (Cross-lingual Prototype Masking), a dynamic and token-wise masking scheme, for multilingual pre-training, using a special token $[\mathcal{C}]_{x}$ to replace a random token $x$ in the input sentence. $[\mathcal{C}]_{x}$ is a cross-lingual prototype for $x$ and then forms an explicit cross-lingual forward pass. We instantiate CLPM for the multilingual pre-training phase of UNMT (unsupervised neural machine translation), and experiments show that CLPM can consistently improve the performance of UNMT models on $\{De, Ro, Ne \} \leftrightarrow En$. Beyond UNMT or bilingual tasks, we show that CLPM can consistently improve the performance of multilingual models on cross-lingual classification."
}