Findings of the 2017 Conference on Machine Translation (WMT17)
Citations
1,219 citations
867 citations
378 citations
Additional excerpts
...We consider data for 13 languages (Ondrej et al., 2017; Bojar et al., 2018; Barrault et al., 2019)....
[...]
319 citations
Cites background from "Findings of the 2017 Conference on ..."
...For DA, it is common practice to convert scores into relative rankings (DARR) when the number of annotations per segment is limited (Bojar et al., 2017b; Ma et al., 2018, 2019)....
[...]
...Reference-less MT evaluation, also known as Quality Estimation (QE), has historically often regressed on HTER for segment-level evaluation (Bojar et al., 2013, 2014, 2015, 2016, 2017a)....
[...]
299 citations
Additional excerpts
...…al., 2014; Bahdanau et al., 2014) have been widely adopted as the stateof-the-art approach for machine translation, both in the research community (Bojar et al., 2016a, 2017, 2018b) and for large-scale production systems (Wu et al., 2016; Zhou et al., 2016; Crego et al., 2016; Hassan et al.,…...
[...]
References
21,126 citations
20,027 citations
"Findings of the 2017 Conference on ..." refers methods in this paper
...The neural system is trained on a bidirectional (forward-backward) RNN-based encoderdecoder30 MT model (Bahdanau et al., 2014) trained for mt → pe translation....
[...]
6,898 citations
"Findings of the 2017 Conference on ..." refers methods in this paper
...(2014) which uses byte-pair encoding (Sennrich et al., 2015) for generating translation tokens....
[...]
4,904 citations
"Findings of the 2017 Conference on ..." refers background in this paper
..., 2013) and a statistical language model (Stolcke, 2002)....
[...]
3,259 citations
"Findings of the 2017 Conference on ..." refers methods in this paper
...Finally, the system was tuned on the development set, optimizing TER/BLEU with Minimum Error Rate Training (Och, 2003)....
[...]
...NMT systems using different input representations are ensembled together in a log-linear model which is tuned for the F1-mult metric using MERT (Och, 2003)....
[...]