You have already added 0 works in your ORCID record related to the merged Research product.
You have already added 0 works in your ORCID record related to the merged Research product.
Data Augmentation for Low-Resource Neural Machine Translation
Data Augmentation for Low-Resource Neural Machine Translation
The quality of a Neural Machine Translation system depends substantially on the availability of sizable parallel corpora. For low-resource language pairs this is not the case, resulting in poor translation quality. Inspired by work in computer vision, we propose a novel data augmentation approach that targets low-frequency words by generating new sentence pairs containing rare words in new, synthetically created contexts. Experimental results on simulated low-resource settings show that our method improves translation quality by up to 2.9 BLEU points over the baseline and up to 3.2 BLEU over back-translation.
- University of Tehran Iran (Islamic Republic of)
- University of Amsterdam Netherlands
- University of Amsterdam (UvA) Netherlands
Microsoft Academic Graph classification: Machine translation Low resource Computer science business.industry media_common.quotation_subject Translation (geometry) computer.software_genre Quality (business) Artificial intelligence business computer Sentence Natural language processing BLEU media_common
Microsoft Academic Graph classification: Machine translation Low resource Computer science business.industry media_common.quotation_subject Translation (geometry) computer.software_genre Quality (business) Artificial intelligence business computer Sentence Natural language processing BLEU media_common
citations This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).175 popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.Top 1% influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).Top 1% impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.Top 1% citations This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).175 popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.Top 1% influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).Top 1% impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.Top 1%
Citations provided by BIP!Popularity provided by BIP!- University of Tehran Iran (Islamic Republic of)
- University of Amsterdam Netherlands
- University of Amsterdam (UvA) Netherlands
The quality of a Neural Machine Translation system depends substantially on the availability of sizable parallel corpora. For low-resource language pairs this is not the case, resulting in poor translation quality. Inspired by work in computer vision, we propose a novel data augmentation approach that targets low-frequency words by generating new sentence pairs containing rare words in new, synthetically created contexts. Experimental results on simulated low-resource settings show that our method improves translation quality by up to 2.9 BLEU points over the baseline and up to 3.2 BLEU over back-translation.