Meaningless yet meaningful: Morphology grounded subword-level NMT

Banerjee, Tamali; Bhattacharyya, Pushpak

doi:10.18653/v1/w18-1207

Cited by 48 publications

(27 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This could be one of the reasons why the BPE-based NMT model is found to be underperforming in this translation task. This finding is corroborated by Banerjee and Bhattacharyya[55] who in their work found that the Morfessor-based segmentation can yield better translation quality than the BPE-based segmentation for linguistically distant language-pairs, and other way round for the close language-pairs.…”

supporting

confidence: 65%

Comparing Statistical and Neural Machine Translation Performance on Hindi-to-Tamil and English-to-Tamil

Ramesh¹,

Parthasarathy²,

Way³

et al. 2020

Preprint

View full text Add to dashboard Cite

Statistical machine translation (SMT) which was the dominant paradigm in machine translation (MT) research for nearly three decades has recently been superseded by the end-to-end deep learning approaches to MT. Although deep neural models produce state-of-the-art results in many translation tasks, they are found to under-perform on resource-poor scenarios. Despite some success, none of the present-day benchmarks that have tried to overcome this problem can be regarded as a universal solution to the problem of translation of many low-resource languages. In this work, we investigate the performance of phrase-based SMT (PB-SMT) and NMT on two rarely-tested low-resource language-pairs, English-to-Tamil and Hindi-to-Tamil, taking a specialised data domain (software localisation) into consideration. This paper demonstrates our findings including the identification of several issues of the current neural approaches to low-resource domain-specific text translation and rankings of our MT systems via a social media platform-based human evaluation scheme.

show abstract

supporting

confidence: 65%

Comparing Statistical and Neural Machine Translation Performance on Hindi-to-Tamil and English-to-Tamil

Ramesh¹,

Parthasarathy²,

Way³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…This could be one of the reasons why the BPE-based NMT model was found to be underperforming in this translation task. This finding was corroborated by Banerjee and Bhattacharyya [60], who in their work found that the Morfessor-based segmentation could yield better translation quality than the BPE-based segmentation for linguistically distant language pairs, and the other way round for close language pairs.…”

Section: The Bpe Segmentation On the Hindi-to-tamil Translationsupporting

confidence: 59%

Comparing Statistical and Neural Machine Translation Performance on Hindi-To-Tamil and English-To-Tamil

Ramesh

Parthasarathy

Haque

et al. 2021

Digital

View full text Add to dashboard Cite

Phrase-based statistical machine translation (PB-SMT) has been the dominant paradigm in machine translation (MT) research for more than two decades. Deep neural MT models have been producing state-of-the-art performance across many translation tasks for four to five years. To put it another way, neural MT (NMT) took the place of PB-SMT a few years back and currently represents the state-of-the-art in MT research. Translation to or from under-resourced languages has been historically seen as a challenging task. Despite producing state-of-the-art results in many translation tasks, NMT still poses many problems such as performing poorly for many low-resource language pairs mainly because of its learning task’s data-demanding nature. MT researchers have been trying to address this problem via various techniques, e.g., exploiting source- and/or target-side monolingual data for training, augmenting bilingual training data, and transfer learning. Despite some success, none of the present-day benchmarks have entirely overcome the problem of translation in low-resource scenarios for many languages. In this work, we investigate the performance of PB-SMT and NMT on two rarely tested under-resourced language pairs, English-to-Tamil and Hindi-to-Tamil, taking a specialised data domain into consideration. This paper demonstrates our findings and presents results showing the rankings of our MT systems produced via a social media-based human evaluation scheme.

show abstract

“…However, these three papers only apply segmentation on the string level and cannot properly handle fusional morphology. Addressing morphology in NMT, Banerjee and Bhattacharyya (2018) combine BPE with a morphological analyzer to "guide" the segmentation of surface forms into substrings. Their approach does not result in morphemes, for example googling → googl|ing, which does not match with google, while in our work we match such morphemes.…”

Section: Related Workmentioning

confidence: 99%

Modeling Word Formation in English–German Neural Machine Translation

Marco¹,

Fraser²

2020

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

View full text Add to dashboard Cite

This paper studies strategies to model word formation in NMT using rich linguistic information, namely a word segmentation approach that goes beyond splitting into substrings by considering fusional morphology. Our linguistically sound segmentation is combined with a method for target-side inflection to accommodate modeling word formation. The best system variants employ source-side morphological analysis and model complex target-side words, improving over a standard system.

show abstract

Meaningless yet meaningful: Morphology grounded subword-level NMT

Cited by 48 publications

References 9 publications

Comparing Statistical and Neural Machine Translation Performance on Hindi-to-Tamil and English-to-Tamil

Comparing Statistical and Neural Machine Translation Performance on Hindi-to-Tamil and English-to-Tamil

Comparing Statistical and Neural Machine Translation Performance on Hindi-To-Tamil and English-To-Tamil

Modeling Word Formation in English–German Neural Machine Translation

Contact Info

Product

Resources

About