2022
DOI: 10.48550/arxiv.2201.06876
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Syntax-based data augmentation for Hungarian-English machine translation

Abstract: We train Transformer-based neural machine translation models for Hungarian-English and English-Hungarian using the Hunglish2 corpus. Our best models achieve a BLEU score of 40.0 on Hungarian-English and 33.4 on English-Hungarian. Furthermore, we present results on an ongoing work about syntax-based augmentation for neural machine translation. Both our code and models are publicly available 1 .

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
2

Relationship

0
2

Authors

Journals

citations
Cited by 2 publications
(2 citation statements)
references
References 18 publications
0
2
0
Order By: Relevance
“…Automated evaluation metrics BLEU is used to assess the quality of the translations. 60 The similarity scores employed for retrieving match sentences are edit distance, fuzzy matching score, n-gram matching score, sentence embedding score. BLEU is measured for these NFR methods.…”
Section: Validation With Dataset Imentioning
confidence: 99%
“…Automated evaluation metrics BLEU is used to assess the quality of the translations. 60 The similarity scores employed for retrieving match sentences are edit distance, fuzzy matching score, n-gram matching score, sentence embedding score. BLEU is measured for these NFR methods.…”
Section: Validation With Dataset Imentioning
confidence: 99%
“…Syntactically aware models have also effectively solved various NLP problems, i.e., sentiment analysis, text generation [88], [89], question answering [90], and semantic role labeling [90], etc. In particular, dependency trees are used as syntactic information in problems such as machine translation [91], [92], language model [93], [94], and semantic 1 https://sentic.net/ Last visited: 04-05-2023 role labeling [95]. Bastings et al [91] presented Graph Convolutional Network (GCN) using a dependency tree for machine translation.…”
Section: Literature Reviewmentioning
confidence: 99%