Results of the WMT17 Neural MT Training Task

Bojar, Ondřej; Helcl, Jindřich; Kocmi, Tom; Libovický, Jindřich; Musil, Tomáš

doi:10.18653/v1/w17-4757

Cited by 20 publications

(23 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use the same model parameters as defined for the WMT 2017 NMT Training Task (Bojar et al, 2017). The task defines models of two sizes, one that fits a 4GB GPU and one that fits an 8GB GPU.…”

Section: Model Detailsmentioning

confidence: 99%

Curriculum Learning and Minibatch Bucketing in Neural Machine Translation

Kocmi

Bojar

2017

RANLP 2017 - Recent Advances in Natural Language Processing Meet Deep Learning

Self Cite

125

124

View full text Add to dashboard Cite

We examine the effects of particular orderings of sentence pairs on the on-line training of neural machine translation (NMT). We focus on two types of such orderings: (1) ensuring that each minibatch contains sentences similar in some aspect and (2) gradual inclusion of some sentence types as the training progresses (so called "curriculum learning"). In our English-to-Czech experiments, the internal homogeneity of minibatches has no effect on the training but some of our "curricula" achieve a small improvement over the baseline.

show abstract

Section: Model Detailsmentioning

confidence: 99%

Curriculum Learning and Minibatch Bucketing in Neural Machine Translation

Kocmi

Bojar

2017

RANLP 2017 - Recent Advances in Natural Language Processing Meet Deep Learning

Self Cite

125

124

View full text Add to dashboard Cite

show abstract

“…Training and development corpora were used from the WMT 2017 shared tasks 11 (Bojar et al, 2017a). Neural Monkey (Helcl and Libovický, 2017), was used to train the NMT systems with configuration provided by the WMT Neural MT Training Task 12 .…”

Section: Finding Correlation Between Neural Network Attention and Outmentioning

confidence: 99%

Hybrid Machine Translation by Combining Output from Multiple Machine Translation Systems

Rikters¹

2019

BJMC

View full text Add to dashboard Cite

This paper aims to combine output from various machine translation (MT) systems so that the overall translation quality of the source text would increase. Applicability of the developed methods for small, morphologically rich and under-resourced languages is evaluated, especially Latvian and Estonian. Existing methods have been analysed, and several combinations of methods have been proposed. The proposed methods have been implemented and evaluated using automatic and human evaluation. During this research novel methods have been created that structure source language sentences into linguistically motivated fragments and combine them using a character level neural language model; combine neural machine translation output by employing sourcetranslation attention alignments; use a multi-pass approach to produce additional incrementally improving training data. The key results of this research are new state-of-the-art machine translation systems for English ↔ Estonian; approaches for utilising neural MT generated attention alignments for MT combination and comprehension of resulting translations; MT combination systems for combining output from English → Latvian statistical MT. A practical application of the methods is implemented and described.

show abstract

“…Machine learning offers a systematic approach to integrate the scores of stand-alone metrics. In the MT evaluation, various successful learning paradigms have been proposed (Bojar et al, 2016), (Bojar et al, 2017) and the existing learning-based metrics can be categorized as binary functions-"which classify the candidate translation as good or bad" (Kulesza and Shieber, 2004), (Guzmán et al, 2015) or continuous functions-"which score the quality of translation on an absolute scale" (Song and Cohn, 2011), (Albrecht and Hwa, 2008). Our research is conceptually similar to the work in (Kulesza and Shieber, 2004), which induces a "human-likeness" criteria.…”

Section: Literature Reviewmentioning

confidence: 99%

Learning-based Composite Metrics for Improved Caption Evaluation

Sharif¹,

White²,

Bennamoun³

et al. 2018

Proceedings of ACL 2018, Student Research Workshop

View full text Add to dashboard Cite

The evaluation of image caption quality is a challenging task, which requires the assessment of two main aspects in a caption: adequacy and fluency. These quality aspects can be judged using a combination of several linguistic features. However, most of the current image captioning metrics focus only on specific linguistic facets, such as the lexical or semantic, and fail to meet a satisfactory level of correlation with human judgements at the sentence-level. We propose a learning-based framework to incorporate the scores of a set of lexical and semantic metrics as features, to capture the adequacy and fluency of captions at different linguistic levels. Our experimental results demonstrate that composite metrics draw upon the strengths of standalone measures to yield improved correlation and accuracy.

show abstract

Results of the WMT17 Neural MT Training Task

Cited by 20 publications

References 25 publications

Curriculum Learning and Minibatch Bucketing in Neural Machine Translation

Curriculum Learning and Minibatch Bucketing in Neural Machine Translation

Hybrid Machine Translation by Combining Output from Multiple Machine Translation Systems

Learning-based Composite Metrics for Improved Caption Evaluation

Contact Info

Product

Resources

About