Investigations on large-scale lightly-supervised training for statistical machine translation.

By Holger Schwenk. 2008

92 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Improving Neural Machine Translation Models with Monolingual Data (2015, methodology): “In phrase-based SMT, we ﬁnd that the use of back-translated training data has a moderate positive effect on the WMT test sets (+0.7 B LEU ), but not on the IWSLT test sets.”
Beyond English-Centric Multilingual Machine Translation (2020, background): “Backtranslation (BT) creates synthetic bitexts from unaligned monolingual data (Schwenk, 2008; Bojar and Tamchyna, 2011; Sennrich et al., 2016a; Edunov et al., 2018; Hoang et al., 2018).”
Investigations on Translation Model Adaptation Using Monolingual Data (2011, methodology): “A somehow related approach was named lightly-supervised training (Schwenk, 2008).”
LIUM’s SMT Machine Translation Systems for WMT 2011 (2011, background): “The resulting bitext contain no new translations, since all words of the translation output come from the translation model, but it contains new combinations (phrases) of known words, and reinforces the probability of some phrase pairs (Schwenk, 2008).”
Neural Machine Translation: A Review (2019, background): “The most successful strategy is called backtranslation (Schwenk, 2008; Sennrich et al., 2016b) which employs a separate translation system in the reverse direction to generate the source sentences for the monolingual target”
Learning Hidden Unit Contribution for Adapting Neural Machine Translation Models (2018, background): “Domain adaptation for phrase-based and related approaches to machine translation has been extensively investigated, e.g. (Schwenk, 2008; Ax-elrod et al., 2011; Carpuat et al., 2013).”
The Edinburgh/LMU Hierarchical Machine Translation System for WMT 2016 (2016, methodology): “• Lightly-supervised training (Schwenk, 2008; Schwenk and Senellart, 2009; Huck et al., 2011).”
LIUM’s SMT Machine Translation Systems for WMT 2012 (2012, background): “The resulting bitexts contain no new translations, since all words of the translation output come from the translation model, but they contain new combinations (phrases) of known words, and reinforce the probability of some phrase pairs (Schwenk, 2008).”
Domain Adaptation for Statistical Machine Translation with Monolingual Resources (2009, methodology): “The adaptation of the translation and re-ordering models is performed by generating synthetic bilingual data from monolingual texts, similarly to what proposed in (Schwenk, 2008).”
Simple Fusion: Return of the Language Model (2018, background): “The source sentences can be generated with a separate translation system (Schwenk, 2008; Sennrich et al., 2016a) (backtranslation), or simply copied over from the target side (Currey et al.”
Improving Gender Translation Accuracy with Filtered Self-Training (2021, unknown): “Self-Training for MT: Monolingual data has been exploited via self-training methods to enhance statistical MT (Schwenk, 2008; Uefﬁng, 2006) as well as neural MT (Wu et al., 2019) by either forward translation of source monolingual data (Ima-mura and Sumita, 2018; Zhang and Zong, 2016) or…”