Machine Translation
Machine translation is the task of translating a sentence in a source language to a different target language.
Approaches for machine translation can range from rule-based to statistical to neural-based. More recently, encoder-decoder attention-based architectures like BERT have attained major improvements in machine translation.
One of the most popular datasets used to benchmark machine translation systems is the WMT family of datasets. Some of the most commonly used evaluation metrics for machine translation systems include BLEU, METEOR, NIST, and others.
( Image credit: Google seq2seq )
Papers
Showing 1–10 of 10752 papers
All datasetsWMT2014 English-GermanWMT2014 English-FrenchIWSLT2014 German-EnglishACESWMT2016 English-RomanianWMT2016 Romanian-EnglishWMT2014 German-EnglishIWSLT2015 German-EnglishWMT2016 English-GermanIWSLT2015 English-VietnameseIWSLT2015 English-GermanWMT2016 German-English
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Exploiting Mono at Scale (single) | SacreBLEU | 40.9 | — | Unverified |
| 2 | MADL | BLEU score | 40.68 | — | Unverified |
| 3 | Attentional encoder-decoder + BPE | BLEU score | 34.2 | — | Unverified |
| 4 | Linguistic Input Features | BLEU score | 28.4 | — | Unverified |
| 5 | DeLighT | BLEU score | 28 | — | Unverified |
| 6 | FLAN 137B (zero-shot) | BLEU score | 27 | — | Unverified |
| 7 | Transformer | BLEU score | 26.7 | — | Unverified |
| 8 | FLAN 137B (few-shot, k=11) | BLEU score | 26.1 | — | Unverified |
| 9 | BiRNN + GCN (Syn + Sem) | BLEU score | 24.9 | — | Unverified |
| 10 | SMT + iterative backtranslation (unsupervised) | BLEU score | 18.23 | — | Unverified |