融合XLM词语表示的神经机器译文自动评价方法(Neural Automatic Evaluation of Machine Translation Method Combined with XLM Word Representation)

Wei Hu (胡纬), Maoxi Li (李茂西), Bailian Qiu (裘白莲), Mingwen Wang (王明文)


Abstract
机器译文自动评价对机器翻译的发展和应用起着重要的促进作用,它一般通过计算机器译文和人工参考译文的相似度来度量机器译文的质量。该文通过跨语种预训练语言模型XLM将源语言句子、机器译文和人工参考译文映射到相同的语义空间,结合分层注意力和内部注意力提取源语言句子与机器译文、机器译文与人工参考译文以及源语言句子与人工参考译文之间差异特征,并将其融入到基于Bi-LSTM神经译文自动评价方法中。在WMT’19译文自动评价数据集上的实验结果表明,融合XLM词语表示的神经机器译文自动评价方法显著提高了其与人工评价的相关性。
Anthology ID:
2021.ccl-1.2
Volume:
Proceedings of the 20th Chinese National Conference on Computational Linguistics
Month:
August
Year:
2021
Address:
Huhhot, China
Editors:
Sheng Li (李生), Maosong Sun (孙茂松), Yang Liu (刘洋), Hua Wu (吴华), Kang Liu (刘康), Wanxiang Che (车万翔), Shizhu He (何世柱), Gaoqi Rao (饶高琦)
Venue:
CCL
SIG:
Publisher:
Chinese Information Processing Society of China
Note:
Pages:
13–22
Language:
Chinese
URL:
https://aclanthology.org/2021.ccl-1.2
DOI:
Bibkey:
Cite (ACL):
Wei Hu, Maoxi Li, Bailian Qiu, and Mingwen Wang. 2021. 融合XLM词语表示的神经机器译文自动评价方法(Neural Automatic Evaluation of Machine Translation Method Combined with XLM Word Representation). In Proceedings of the 20th Chinese National Conference on Computational Linguistics, pages 13–22, Huhhot, China. Chinese Information Processing Society of China.
Cite (Informal):
融合XLM词语表示的神经机器译文自动评价方法(Neural Automatic Evaluation of Machine Translation Method Combined with XLM Word Representation) (Hu et al., CCL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.ccl-1.2.pdf