Shuheng Wang , Heyan Huang , Shumin Shi , Dongbai Li , Dongen Guo
{"title":"Correcting translation for non-autoregressive transformer","authors":"Shuheng Wang , Heyan Huang , Shumin Shi , Dongbai Li , Dongen Guo","doi":"10.1016/j.asoc.2024.112488","DOIUrl":null,"url":null,"abstract":"<div><div>Non-Autoregressive Transformer has shown great success in recent years. It generally employs the encoder–decoder framework, where the encoder maps the sentence into hidden representation, and the decoder generates the target tokens simultaneously. Since the theory of non-autoregressive transformer is consistent with the architecture of the encoder, we suppose that it is somewhat wasteful for the encoder to only map input sentence into hidden representation. In this study, we proposed a novel non-autoregressive transformer to fully exploit the capabilities of the encoder. Specifically, in our approach, the encoder not only encodes the input sentence into hidden representation, but also generates the target tokens. Consequently, the decoder is relieved of its responsibility to generate the target tokens, instead of focusing on correcting the sentence produced by the encoder. We evaluate the performance of the proposed non-autoregressive transformer on three widely-used translation tasks. The experimental results illustrate the proposed method can significantly improve the performance of the non-autoregressive transformer , which achieved 27.94 BLEU on WMT14 EN <span><math><mo>→</mo></math></span> DE task, 33.96 BLEU on WMT16 EN <span><math><mo>→</mo></math></span> RO task, and 33.85 BLEU on IWSLT14 DE <span><math><mo>→</mo></math></span> EN.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"168 ","pages":"Article 112488"},"PeriodicalIF":7.2000,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Soft Computing","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1568494624012626","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0
Abstract
Non-Autoregressive Transformer has shown great success in recent years. It generally employs the encoder–decoder framework, where the encoder maps the sentence into hidden representation, and the decoder generates the target tokens simultaneously. Since the theory of non-autoregressive transformer is consistent with the architecture of the encoder, we suppose that it is somewhat wasteful for the encoder to only map input sentence into hidden representation. In this study, we proposed a novel non-autoregressive transformer to fully exploit the capabilities of the encoder. Specifically, in our approach, the encoder not only encodes the input sentence into hidden representation, but also generates the target tokens. Consequently, the decoder is relieved of its responsibility to generate the target tokens, instead of focusing on correcting the sentence produced by the encoder. We evaluate the performance of the proposed non-autoregressive transformer on three widely-used translation tasks. The experimental results illustrate the proposed method can significantly improve the performance of the non-autoregressive transformer , which achieved 27.94 BLEU on WMT14 EN DE task, 33.96 BLEU on WMT16 EN RO task, and 33.85 BLEU on IWSLT14 DE EN.
期刊介绍:
Applied Soft Computing is an international journal promoting an integrated view of soft computing to solve real life problems.The focus is to publish the highest quality research in application and convergence of the areas of Fuzzy Logic, Neural Networks, Evolutionary Computing, Rough Sets and other similar techniques to address real world complexities.
Applied Soft Computing is a rolling publication: articles are published as soon as the editor-in-chief has accepted them. Therefore, the web site will continuously be updated with new articles and the publication time will be short.