where hm is a feature function, Am is the associated feature weight , and Z (f) is a constant for normalization:

As our decoder accounts for multiple derivations, we extend the MERT algorithm to tune feature weights with respect to BLEU score for max-translation decoding.

We found that accounting for all possible derivations in max-translation decoding resulted in a small negative effect on BLEU score (from 30.11 to 29.82), even though the feature weights were tuned with respect to BLEU score.

We concatenate and normalize their feature weights for the joint decoder.

Minimum error rate training (Och, 2003) is widely used to optimize feature weights for a linear model (Och and Ney, 2002).

The key idea of MERT is to tune one feature weight to minimize error rate each time while keep others fixed.

where a is the feature value of current dimension, cc is the feature weight being tuned, and b is the dotproduct of other dimensions.

0 As multiple derivations are used for finding optimal translations, we extend the minimum error rate training (MERT) algorithm (Och, 2003) to tune feature weights with respect to BLEU score for max-translation decoding (Section 4).

Let 2m be the feature weight vector for member decoder dm, the training procedure proceeds as follows:

For each decoder dm, find a new feature weight vector 2;,1 which optimizes the specified evaluation criterion L on D using the MERT algorithm based on the n-best list Jim generated by dm:

where T denotes the translations selected by re-ranking the translations in Jim using a new feature weight vector A

Feature Weights .

Finally, we make two observations about feature weights in the RTER model.

Second, good MT evaluation feature weights are not good weights for RTE.