We introduce an approach to optimize a ma- chine translation (MT) system on multiple metrics simultaneously. Different metrics (e.g. BLEU, TER) focus on different aspects of translation quality; our multi-objective ap- proach leverages these diverse aspects to im- prove overall quality. Our approach is based on the theory of Pareto Optimality. It is simple to implement on top of existing single-objective optimization meth- ods (e.g. MERT, PRO) and outperforms ad hoc alternatives based on linear-combination of metrics. We also discuss the issue of metric tunability and show that our Pareto approach is more effective in incorporating new metrics from MT evaluation for MT optimization.

Click Here to View the Paper In XML Format