Paper: Learning to Translate with Multiple Objectives

ACL ID P12-1001
Title Learning to Translate with Multiple Objectives
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2012

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.