在汉语等其他有省略代词习惯的语言中,通常会省略可从上下文信息推断出的代词。尽管以Transformer为代表的的神经机器翻译模型取得了巨大的成功,但这种代词省略现象依旧使神经机器翻译模型面临很大的挑战。该文在Transformer模型基础上提出了一个融合零指代识别的翻译模型,并引入篇章上下文来丰富指代信息。具体地,该模型采用联合学习的框架,在翻译模型基础上,联合了一个分类任务,即判别句子中省略代词在句子所表示的成分,使得模型能够融合零指代信息辅助翻译。通过在中英对话数据集上的实验,验证了该文所提出方法的有效性,与基准模型相比,翻译性能提升了1.48个BLEU值。
Abstract
Pronouns are usually omitted if it can be inferred from context in such pro-drop languages as Chinese, which poses a significant challenge for NMT. This paper proposes a model to capture zero pronoun information based on Transformer, introducing the context to enrich anaphora information. It adopts the framework of joint learning, employing a classification task to identify the omitted component of the sentence to help translation optimization. Experiments on a Chinese-English dialogue dataset prove the proposed method achieves an improvement of 1.48 BLEU compared with the benchmark model.
关键词
零指代 /
篇章级机器翻译 /
联合学习
{{custom_keyword}} /
Key words
zero pronoun /
context-aware machine translation /
joint learning
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] YANG Y Q, LIU Y L, XUE N W, et al. Recovering dropped pronouns from chinese text messages[C]//Proceedings of the ACL, 2015: 309-313.
[2] WANG L Y, TU Z P, SHI S M, et al. Translating pro-drop languages with reconstruction models[C]//Proceedings of the AAAI, 2018: 4937-4945.
[3] WANG L Y, TU Z P, ZHANG X J, et al. A novel approach to dropped pronoun translation[C]//Proceedings of the NAACL, 2016: 983-993.
[4] TAN X, KUANG S H, XIONG D Y, et al. Detecting and translating dropped pronouns in neural machine translation[C]//Proceedings of the NLPCC, 2019: 343-354.
[5] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the NIPS, 2017: 6000-6010.
[6] ZHAO S H,HWEE T N. Identification and resolution of Chinese zero pronouns: A machine learning approach. [C] //Proceedings of the EMNLP, 2007: 541-550.
[7] CHEN C, VINCENT N. Chinese zero pronoun resolution: some recent advances[C]//Proceedings of the EMNLP, 2013: 1360-1365.
[8] XIONG H, HE Z, WU H, et al. Modeling coherence for discourse neural machine translation[C]//Proceedings of the AAAI, 2019: 7338-7345.
[9] ZHANG J, LUAN H, SUN M, et al. Improving the transformer translation model with document-level context[C]//Proceedings of the EMNLP, 2018: 533-542.
[10] BAWDEN R, SENNRICH R, BIRCH A, et al. Evaluating discourse phenomena in neural machine translation[C]//Proceedings of the NAACL, 2018: 1304-1313.
[11] WANG L Y, TU Z P, WANG X, et al.One model to learn both: Zero pronoun prediction and translation[C]//Proceedings of the EMNLP, 2019: 921-930.
[12] WANG L, TU Z, WAY A, et al. Exploiting cross-sentence context for neural machine translation[C]//Proceedings of the EMNLP, 2017: 2826-2831.
[13] MARUF S, MARTINS F T, HAFFARI G. Selective attention for context-aware neural machine translation[C]//Proceedings of the ACL, 2019: 3092-3102.
[14] YANG Z X, ZHANG J C, MENG F D, et al. Enhancing context modeling with a query-guided capsule network for document-level translation[C]//Proceedings of the EMNLP, 2019: 1527-1537.
[15] CHEN L Q, LI J H, GONG Z X, et al. Improving context-aware neural machine translation with source-side monolingual documents[C]//Proceedings of the IJCAI, 2021: 3794-3800.
[16] CHEN L Q, LI J H, GONG Z X, et al. Breaking the corpus bottleneck for context-aware neural machine translation with a novel joint pre-training approach[C]//Proceedings of the ACL, 2021: 2851-2861.
[17] VOITA E, SENNRICH R, TITOV I. When a good translation is wrong in context: Context-aware machine translation improves on deixis, ellipsis, and lexical cohesion[C]//Proceedings of the ACL, 2019a: 1198-1212.
[18] KOEHN P, HOANG H, BIRCH A, et al. Moses: Open source toolkit for statistical machine translation[C]//Proceedings of the ACL, System Demonstrations, 2007: 177-180.
[19] SENNRICH R, HADDOW B, BIRCH A. Neural machine translation of rare words with subword units[C]//Proceedings of the ACL, 2016: 1715-1725.
[20] PAPINENI K, ROUKOS S, WARD T, et al. Bleu: A method for automatic evaluation of machine translation[C]//Proceedings of the ACL, 2002: 311-318.
[21] PHILIPP K. Statistical significance tests for machine translation evaluation[C]//Proceedings of the EMNLP, 2004: 388-395.
[22] KINGMA D P, BA J. Adam: A method for stochastic optimization[C]//Proceedings of the ICLR, 2015.
[23] MICULICICH W L, RAM D, PAPPAS N, et al. Document-level neural machine translation with hierarchical attention networks[C]//Proceedings of the EMNLP, 2018: 2947-2954.
[24] MICULICICH W L,ANDREI PB. Validation of an automatic metric for the accuracy of pronoun translation[C]//Proceedings of the EMNLP, 2017: 17-25.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1975 KB)
