Many machine translation (MT) evaluation metrics have been shown to correlate better with human judgment than BLEU. In principle, tuning on these metrics should yield better syste...
Minimum error rate training (MERT) involves choosing parameter values for a machine translation (MT) system that maximize performance on a tuning set as measured by an automatic e...
Recently, confusion network decoding has been applied in machine translation system combination. Due to errors in the hypothesis alignment, decoding may result in ungrammatical co...
Antti-Veikko I. Rosti, Spyridon Matsoukas, Richard...