Knowledge Commons of Institute of Automation,CAS
CASIA’s System for IWSLT 2020 Open Domain Translation | |
Wang, Qian1,2; Liu, Yuchen1,2; Ma, Cong1,2; Lu, Yu1,2; Wang, Yining1,2; Zhou, Long1,2; Zhao, Yang1,2; Zhang, Jiajun1,2; Zong, Chengqing1,2 | |
2020-06-09 | |
会议名称 | the 17th International Conference on Spoken Language Translation (IWSLT) |
会议日期 | 2020-6-9 |
会议地点 | Online |
摘要 | This paper describes the CASIA’s system for the IWSLT 2020 open domain transla- tion task. This year we participate in both Chinese→Japanese and Japanese→Chinese translation tasks. Our system is neural ma- chine translation system based on Transformer model. We augment the training data with knowledge distillation and back translation to improve the translation performance. Domain data classification and weighted domain model ensemble are introduced to generate the final translation result. We compare and analyze the performance on development data with differ- ent model settings and different data process- ing techniques. |
七大方向——子方向分类 | 自然语言处理 |
国重实验室规划方向分类 | 语音语言处理 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/50875 |
专题 | 多模态人工智能系统全国重点实验室_自然语言处理 |
作者单位 | 1.National Laboratory of Pattern Recognition, CASIA, Beijing, China 2.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China |
第一作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Wang, Qian,Liu, Yuchen,Ma, Cong,等. CASIA’s System for IWSLT 2020 Open Domain Translation[C],2020. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
2020.iwslt-1.15.pdf(547KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论