Showing 1–1 of 1 results for author: ElMosalami, J S

Ensembling of Distilled Models from Multi-task Teachers for Constrained Resource Language Pairs

Authors: Amr Hendy, Esraa A. Gad, Mohamed Abdelghaffar, Jailan S. ElMosalami, Mohamed Afify, Ahmed Y. Tawfik, Hany Hassan Awadalla

Abstract: This paper describes our submission to the constrained track of WMT21 shared news translation task. We focus on the three relatively low resource language pairs Bengali to and from Hindi, English to and from Hausa, and Xhosa to and from Zulu. To overcome the limitation of relatively low parallel data we train a multilingual model using a multitask objective employing both parallel and monolingual… ▽ More This paper describes our submission to the constrained track of WMT21 shared news translation task. We focus on the three relatively low resource language pairs Bengali to and from Hindi, English to and from Hausa, and Xhosa to and from Zulu. To overcome the limitation of relatively low parallel data we train a multilingual model using a multitask objective employing both parallel and monolingual data. In addition, we augment the data using back translation. We also train a bilingual model incorporating back translation and knowledge distillation then combine the two models using sequence-to-sequence mapping. We see around 70% relative gain in BLEU point for English to and from Hausa, and around 25% relative improvements for both Bengali to and from Hindi, and Xhosa to and from Zulu compared to bilingual baselines. △ Less

Submitted 25 November, 2021; originally announced November 2021.