@inproceedings{hongbao-etal-2022-transns,
    title = "Transn{'}s Submissions to the {WMT}22 Translation Suggestion Task",
    author = "Hongbao, Mao  and
      Wenbo, Zhang  and
      Jie, Cai  and
      Jianwei, Cheng",
    editor = {Koehn, Philipp  and
      Barrault, Lo{\"i}c  and
      Bojar, Ond{\v{r}}ej  and
      Bougares, Fethi  and
      Chatterjee, Rajen  and
      Costa-juss{\`a}, Marta R.  and
      Federmann, Christian  and
      Fishel, Mark  and
      Fraser, Alexander  and
      Freitag, Markus  and
      Graham, Yvette  and
      Grundkiewicz, Roman  and
      Guzman, Paco  and
      Haddow, Barry  and
      Huck, Matthias  and
      Jimeno Yepes, Antonio  and
      Kocmi, Tom  and
      Martins, Andr{\'e}  and
      Morishita, Makoto  and
      Monz, Christof  and
      Nagata, Masaaki  and
      Nakazawa, Toshiaki  and
      Negri, Matteo  and
      N{\'e}v{\'e}ol, Aur{\'e}lie  and
      Neves, Mariana  and
      Popel, Martin  and
      Turchi, Marco  and
      Zampieri, Marcos},
    booktitle = "Proceedings of the Seventh Conference on Machine Translation (WMT)",
    month = dec,
    year = "2022",
    address = "Abu Dhabi, United Arab Emirates (Hybrid)",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2022.wmt-1.124/",
    pages = "1205--1210",
    abstract = "This paper describes the Transn{'}s submissions to the WMT2022 shared task on TranslationSuggestion. Our team participated on two tasks: Naive Translation Suggestion and TranslationSuggestion with Hints, focusing on two language directions Zh{\textrightarrow}En and En{\textrightarrow}Zh. Apart from the golden training data provided by the shared task, we utilized synthetic corpus to fine-tune on DeltaLM ({\ensuremath{\Delta}}LM), which is a pre-trained encoder-decoder language model. We applied two-stage training strategy on {\ensuremath{\Delta}}LM and several effective methods to generate synthetic corpus, which contribute a lot to the results. According to the official evaluation results in terms of BLEU scores, our submissions in Naive Translation Suggestion En{\textrightarrow}Zh and Translation Suggestion with Hints (both Zh{\textrightarrow}En and En{\textrightarrow}Zh) ranked 1st, and Naive Translation Suggestion Zh{\textrightarrow}En also achieved comparable result to the best score."
}