中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
Self-Modifying State Modeling for Simultaneous Machine Translation

文献类型:会议论文

作者Donglei, Yu1,2; Xiaomian, Kang1,2; Yuchen, Liu1,2; YU, Zhou1,3; Chengqing, Zong1,2
出版日期2024
会议日期August 11–16, 2024
会议地点Bangkok, Thailand
英文摘要

Simultaneous Machine Translation (SiMT) generates target outputs while receiving stream source inputs and requires a read/write policy to decide whether to wait for the next source token or generate a new target token, whose decisions form a decision path. Existing SiMT methods, which learn the policy by exploring various decision paths in training, face inherent limitations. These methods not only fail to precisely optimize the policy due to the inability to accurately assess the individual impact of each decision on SiMT performance, but also cannot sufficiently explore all potential paths because of their vast number. Besides, building decision paths requires unidirectional encoders to simulate streaming source inputs, which impairs the translation quality of SiMT models. To solve these issues, we propose Self-Modifying State Modeling (SM2), a novel training paradigm for SiMT task. Without building decision paths, SM2 individually optimizes decisions at each state during training. To precisely optimize the policy, SM2 introduces Self-Modifying process to independently assess and adjust decisions at each state. For sufficient exploration, SM2 proposes Prefix Sampling to efficiently traverse all potential states. Moreover, SM2 ensures compatibility with bidirectional encoders, thus achieving higher translation quality. Experiments show that SM2 outperforms strong baselines. Furthermore, SM2 allows offline machine translation models to acquire SiMT ability with fine-tuning.

语种英语
源URL[http://ir.ia.ac.cn/handle/173211/57442]  
专题模式识别国家重点实验室_自然语言处理
通讯作者YU, Zhou
作者单位1.State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences
2.School of Artificial Intelligence, University of Chinese Academy of Sciences
3.Fanyu AI Laboratory, Zhongke Fanyu Technology Co., Ltd
推荐引用方式
GB/T 7714
Donglei, Yu,Xiaomian, Kang,Yuchen, Liu,et al. Self-Modifying State Modeling for Simultaneous Machine Translation[C]. 见:. Bangkok, Thailand. August 11–16, 2024.

入库方式: OAI收割

来源:自动化研究所

浏览0
下载0
收藏0
其他版本

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。