CUNI Submission in WMT22 General Task
This is an incremental improvement for machine translation researchers and practitioners, specifically in the English-to-Czech direction.
The paper tackled improving English-to-Czech machine translation by exploring block backtranslation techniques, showing that combining MBR decoding with mixed backtranslation training yields better results, with improvements measured in COMET score and named entities translation accuracy.
We present the CUNI-Bergamot submission for the WMT22 General translation task. We compete in English$\rightarrow$Czech direction. Our submission further explores block backtranslation techniques. Compared to the previous work, we measure performance in terms of COMET score and named entities translation accuracy. We evaluate performance of MBR decoding compared to traditional mixed backtranslation training and we show a possible synergy when using both of the techniques simultaneously. The results show that both approaches are effective means of improving translation quality and they yield even better results when combined.