View on GitHub

sose20-mt

Machine Translation

Special topics in MT

Read, understand and present scientific papers on current challenges in machine translation. You are expected to choose a topic from the list below and give a short presentation about it (between 15 and 20 min).

Multilingual translation models

Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation
https://www.aclweb.org/anthology/Q17-1024
A Comparison of Transformer and Recurrent Neural Networks on Multilingual Neural Machine Translation
https://aclweb.org/anthology/C18-1054
Multilingual Neural Machine Translation with Task-Specific Attention
https://aclweb.org/anthology/C18-1263

Noisy data

MTNT: A Testbed for Machine Translation of Noisy Text
http://www.cs.cmu.edu/~pmichel1/hosting/mtnt-emnlp.pdf
Synthetic and Natural Noise Both Break Neural Machine Translation
https://arxiv.org/abs/1711.02173 Toward Robust Neural Machine Translation for Noisy Input Sequences
https://pdfs.semanticscholar.org/88ed/f12127a628bed608cae0bdf3700d00824df4.pdf On the Impact of Various Types of Noise on Neural Machine Translation https://aclweb.org/anthology/W18-2709

Domain adaptation

An Empirical Comparison of Domain Adaptation Methods for Neural Machine Translation
https://aclweb.org/anthology/P17-2061
Dynamic Data Selection for Neural Machine Translation
https://arxiv.org/pdf/1708.00712.pdf

Document-level machine translation

Evaluating Discourse Phenomena in Neural Machine Translation
https://arxiv.org/pdf/1711.00513.pdf
Neural Machine Translation with Extended Context
http://www.aclweb.org/anthology/W17-4811
Context-Aware Neural Machine Translation Learns Anaphora Resolution
http://aclweb.org/anthology/P18-1117

Human parity

Achieving Human Parity on Automatic Chinese to English News Translation
https://www.microsoft.com/en-us/research/uploads/prod/2018/03/final-achieving-human.pdf
Attaining the Unattainable? Reassessing Claims of Human Parity in Neural Machine Translation
https://www.aclweb.org/anthology/W18-6312
Has Machine Translation Achieved Human Parity? A Case for Document-level Evaluation
http://aclweb.org/anthology/D18-1512

Translation of literature

What Level of Quality can Neural Machine Translation Attain on Literary Text?
https://arxiv.org/pdf/1801.04962.pdf
Ethical issues regarding machine(-assisted) translation of literary texts
https://www.tandfonline.com/doi/pdf/10.1080/0907676X.2018.1520907?needAccess=true

Multimodal machine translation (Speech)

Towards speech-to-text translation without speech recognition
https://aclweb.org/anthology/E17-2076
Sequence-to-Sequence Models Can Directly Translate Foreign Speech
https://arxiv.org/pdf/1703.08581.pdf

Multimodal machine translation (Images)

A Shared Task on Multimodal Machine Translation and Crosslingual Image Description
https://aclweb.org/anthology/W16-2346
Adversarial Evaluation of Multimodal Machine Translation
https://www.aclweb.org/anthology/D18-1329

Low-resource scenarios

Rapid Adaptation of Neural Machine Translation to New Languages
https://aclweb.org/anthology/D18-1103
Copied Monolingual Data Improves Low-Resource Neural Machine Translation
http://aclweb.org/anthology/W17-4715
Transfer Learning for Low-Resource Neural Machine Translation
https://www.isi.edu/natural-language/mt/emnlp16-transfer.pdf

Evaluation of linguistic phenomena

How Grammatical is Character-level Neural Machine Translation? Assessing MT Quality with Contrastive Translation Pairs
http://aclweb.org/anthology/E17-2060
Evaluating the morphological competence of Machine Translation Systems
https://www.aclweb.org/anthology/W17-4705
What do Neural Machine Translation Models Learn about Morphology?
http://aclweb.org/anthology/P17-1080

General purpose sentence representations

Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning
https://arxiv.org/pdf/1804.00079.pdf
Learning Joint Multilingual Sentence Representations with Neural Machine Translation
https://research.fb.com/wp-content/uploads/2017/08/multiling_repl4nlp17_fair.pdf

NMT for creative language generation

A Master-Apprentice Approach to Automatic Creation of Culturally Satirical Movie Titles
https://www.aclweb.org/anthology/W18-6534
Neural Poetry Translation
https://aclweb.org/anthology/N18-2011
Neural Melody Composition from Lyrics
https://arxiv.org/pdf/1809.04318.pdf

Word segmentation

Fully Character-Level Neural Machine Translation without Explicit Segmentation
http://aclweb.org/anthology/Q17-1026
SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing
https://arxiv.org/pdf/1808.06226.pdf
Cognate-aware morphological segmentation for multilingual neural translation
https://www.aclweb.org/anthology/W18-6410

Unsupervised machine translation

Unsupervised Neural Machine Translation
https://openreview.net/pdf?id=Sy2ogebAW
Unsupervised Machine Translation Using Monolingual Corpora Only
https://openreview.net/pdf?id=rkYTTf-AZ
Phrase-Based & Neural Unsupervised Machine Translation
https://aclweb.org/anthology/D18-1549

Dealing with long sentences

Overcoming the Curse of Sentence Length for Neural Machine Translation using Automatic Segmentation
https://arxiv.org/pdf/1409.1257.pdf

Also worth checking out:

The following list contains additional references on these and other topics:
https://github.com/THUNLP-MT/MT-Reading-List