Dynamic Context Selection for Document-level Neural Machine Translation via Reinforcement Learning | |
Kang, Xiaomian1,2; Zhao, Yang1,2; Zhang, Jiajun1,2,3; Zong, Chengqing1,2,4 | |
2020-11 | |
会议日期 | November 16–20, 2020 |
会议地点 | Online |
关键词 | Docment-level NMT Neural Machine Translation Reinforcement Learning Context Selection |
英文摘要 | Document-level neural machine translation has yielded attractive improvements. However, majority of existing methods roughly use all context sentences in a fixed scope. They neglect the fact that different source sentences need different sizes of context. To address this problem, we propose an effective approach to select dynamic context so that the document-level translation model can utilize the more useful selected context sentences to produce better translations. Specifically, we introduce a selection module that is independent of the translation module to score each candidate context sentence. Then, we propose two strategies to explicitly select a variable number of context sentences and feed them into the translation module. We train the two modules end-to-end via reinforcement learning. A novel reward is proposed to encourage the selection and utilization of dynamic context sentences. Experiments demonstrate that our approach can select adaptive context sentences for different source sentences, and significantly improves the performance of document-level translation methods. |
语种 | 英语 |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/44305] |
专题 | 模式识别国家重点实验室_自然语言处理 |
作者单位 | 1.National Laboratory of Pattern Recognition, Institute of Automation, CAS, Beijing, China 2.Beijing Academy of Artificial Intelligence, Beijing, China 3.CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing, China 4.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China |
推荐引用方式 GB/T 7714 | Kang, Xiaomian,Zhao, Yang,Zhang, Jiajun,et al. Dynamic Context Selection for Document-level Neural Machine Translation via Reinforcement Learning[C]. 见:. Online. November 16–20, 2020. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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