Touch Editing: A Flexible One-Time Interaction Approach for Translation | |
Wang, Qian2,3; Zhang, Jiajun2,3; Liu, Lemao1; Huang, Guoping1; Zong, Chengqing2,3 | |
2020-12-04 | |
会议日期 | 2020-12-4 |
会议地点 | Suzhou, China |
英文摘要 | We propose a touch-based editing method for translation, which is more flexible than tradi- tional keyboard-mouse-based translation post- editing. This approach relies on touch actions that users perform to indicate translation errors. We present a dual-encoder model to handle the actions and generate refined translations. To mimic the user feedback, we adopt the TER al- gorithm comparing between draft translations and references to automatically extract the sim- ulated actions for training data construction. Experiments on translation datasets with sim- ulated editing actions show that our method significantly improves original translation of Transformer (up to 25.31 BLEU) and outper- forms existing interactive translation methods (up to 16.64 BLEU). We also conduct ex- periments on post-editing dataset to further prove the robustness and effectiveness of our method. |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/50874] ![]() |
专题 | 模式识别国家重点实验室_自然语言处理 |
作者单位 | 1.Tencent AI Lab, Shenzhen, China 2.2. University of Chinese Academy of Sciences, Beijing, China 3.1. National Laboratory of Pattern Recognition, CASIA, Beijing, China |
推荐引用方式 GB/T 7714 | Wang, Qian,Zhang, Jiajun,Liu, Lemao,et al. Touch Editing: A Flexible One-Time Interaction Approach for Translation[C]. 见:. Suzhou, China. 2020-12-4. |
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