Towards Unified Multi-Domain Machine Translation With Mixture of Domain Experts
文献类型：期刊论文
| 作者 | Lu, Jinliang2,3 ; Zhang, Jiajun1,2,3
|
| 刊名 | IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
 |
| 出版日期 | 2023 |
| 卷号 | 31页码:3488-3498 |
| 关键词 | Training Adaptation models Transformers Task analysis Speech processing Machine translation Switches Machine Translation Multi-domain Mixture-of-expert |
| ISSN号 | 2329-9290 |
| DOI | 10.1109/TASLP.2023.3316451 |
| 通讯作者 | Zhang, Jiajun(jjzhang@nlpr.ia.ac.cn) |
| 英文摘要 | machine translation (MDMT) aims to construct models with mixed-domain training corpora to switch translation between different domains. Previous studies either assume that the domain information is given and leverage the domain knowledge to guide the translation process, or suppose that the domain information is unknown and utilize the model to automatically recognize it. However, the cases are mixed in practical scenarios, which means that some sentences are labeled with domain information while others are unlabeled, which is beyond the capacity of the previous methods. In this article, we propose a unified MDMT model with a mixture of sub-networks (experts) to address the cases with or without domain labels. The mixture of sub-networks in our MDMT model includes a shared expert and multiple domain-specific experts. For the inputs with domain labels, our MDMT model goes through the shared and the corresponding domain-specific experts. For the unlabeled inputs, our MDMT model activates all the experts, each of which makes a dynamic contribution. Experimental results on multiple diverse domains in De -> En, Fr--> En, and En -> Ro demonstrate that our method can outperform the strong baselines in both scenarios with or without domain labels. Further analyses show that our model has good generalization ability when transferring into new domains. |
| 资助项目 | National Key R&D Program of China[2022ZD0160602] ; Natural Science Foundation of China[62122088] |
| WOS研究方向 | Acoustics ; Engineering |
| 语种 | 英语 |
| WOS记录号 | WOS:001089305500009 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 资助机构 | National Key R&D Program of China ; Natural Science Foundation of China |
| 源URL | [http://ir.ia.ac.cn/handle/173211/54439]  |
| 专题 | 紫东太初大模型研究中心 模式识别国家重点实验室_自然语言处理 |
| 通讯作者 | Zhang, Jiajun |
| 作者单位 | 1.Wuhan AI Res, Wuhan 430072, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 3.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China |
| 推荐引用方式 GB/T 7714 | Lu, Jinliang,Zhang, Jiajun. Towards Unified Multi-Domain Machine Translation With Mixture of Domain Experts[J]. IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING,2023,31:3488-3498. |
| APA | Lu, Jinliang,&Zhang, Jiajun.(2023).Towards Unified Multi-Domain Machine Translation With Mixture of Domain Experts.IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING,31,3488-3498. |
| MLA | Lu, Jinliang,et al."Towards Unified Multi-Domain Machine Translation With Mixture of Domain Experts".IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING 31(2023):3488-3498. |
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。