%0 Conference Proceedings %T The NiuTrans Machine Translation System for WMT18 %A Wang, Qiang %A Li, Bei %A Liu, Jiqiang %A Jiang, Bojian %A Zhang, Zheyang %A Li, Yinqiao %A Lin, Ye %A Xiao, Tong %A Zhu, Jingbo %Y Bojar, Ondřej %Y Chatterjee, Rajen %Y Federmann, Christian %Y Fishel, Mark %Y Graham, Yvette %Y Haddow, Barry %Y Huck, Matthias %Y Yepes, Antonio Jimeno %Y Koehn, Philipp %Y Monz, Christof %Y Negri, Matteo %Y Névéol, Aurélie %Y Neves, Mariana %Y Post, Matt %Y Specia, Lucia %Y Turchi, Marco %Y Verspoor, Karin %S Proceedings of the Third Conference on Machine Translation: Shared Task Papers %D 2018 %8 October %I Association for Computational Linguistics %C Belgium, Brussels %F wang-etal-2018-niutrans %X This paper describes the submission of the NiuTrans neural machine translation system for the WMT 2018 Chinese ↔ English news translation tasks. Our baseline systems are based on the Transformer architecture. We further improve the translation performance 2.4-2.6 BLEU points from four aspects, including architectural improvements, diverse ensemble decoding, reranking, and post-processing. Among constrained submissions, we rank 2nd out of 16 submitted systems on Chinese → English task and 3rd out of 16 on English → Chinese task, respectively. %R 10.18653/v1/W18-6430 %U https://aclanthology.org/W18-6430 %U https://doi.org/10.18653/v1/W18-6430 %P 528-534