Improving neural machine translation with conditional sequence generative adversarial nets

Z Yang, W Chen, F Wang, B Xu - arxiv preprint arxiv:1703.04887, 2017‏ - arxiv.org
This paper proposes an approach for applying GANs to NMT. We build a conditional
sequence generative adversarial net which comprises of two adversarial sub models, a …

deltaBLEU: A discriminative metric for generation tasks with intrinsically diverse targets

M Galley, C Brockett, A Sordoni, Y Ji, M Auli… - arxiv preprint arxiv …, 2015‏ - arxiv.org
We introduce Discriminative BLEU (deltaBLEU), a novel metric for intrinsic evaluation of
generated text in tasks that admit a diverse range of possible outputs. Reference strings are …

Learning paraphrastic sentence embeddings from back-translated bitext

J Wieting, J Mallinson, K Gimpel - arxiv preprint arxiv:1706.01847, 2017‏ - arxiv.org
We consider the problem of learning general-purpose, paraphrastic sentence embeddings
in the setting of Wieting et al.(2016b). We use neural machine translation to generate …

[PDF][PDF] The AMARA Corpus: Building Parallel Language Resources for the Educational Domain.

A Abdelali, F Guzman, H Sajjad, S Vogel - LREC, 2014‏ - academia.edu
This paper presents the AMARA corpus of on-line educational content: a new parallel
corpus of educational video subtitles, multilingually aligned for 20 languages, ie 20 …

Challenges of neural machine translation for short texts

Y Wan, B Yang, DF Wong, LS Chao, L Yao… - Computational …, 2022‏ - direct.mit.edu
Short texts (STs) present in a variety of scenarios, including query, dialog, and entity names.
Most of the exciting studies in neural machine translation (NMT) are focused on tackling …

It's MBR All the Way Down: Modern Generation Techniques Through the Lens of Minimum Bayes Risk

A Bertsch, A **e, G Neubig, MR Gormley - arxiv preprint arxiv:2310.01387, 2023‏ - arxiv.org
Minimum Bayes Risk (MBR) decoding is a method for choosing the outputs of a machine
learning system based not on the output with the highest probability, but the output with the …

End-to-end speech translation with transcoding by multi-task learning for distant language pairs

T Kano, S Sakti, S Nakamura - IEEE/ACM Transactions on …, 2020‏ - ieeexplore.ieee.org
Directly translating spoken utterances from a source language to a target language is
challenging because it requires a fundamental transformation in both linguistic and para/non …

An automatic evaluation metric for Ancient-Modern Chinese translation

K Yang, D Liu, Q Qu, Y Sang, J Lv - Neural Computing and Applications, 2021‏ - Springer
As a written language used for thousands of years, Ancient Chinese has some special
characteristics like complex semantics as polysemy and the one-to-many alignment with …

A user-study on online adaptation of neural machine translation to human post-edits

S Karimova, P Simianer, S Riezler - Machine Translation, 2018‏ - Springer
The advantages of neural machine translation (NMT) have been extensively validated for
offline translation of several language pairs for different domains of spoken and written …

[PDF][PDF] Phrasal: A toolkit for new directions in statistical machine translation

S Green, D Cer, CD Manning - … of the ninth workshop on statistical …, 2014‏ - aclanthology.org
We present a new version of Phrasal, an open-source toolkit for statistical phrasebased
machine translation. This revision includes features that support emerging research trends …