Abstract

Patent translation is a complex problem due to the highly specialized technical vocabulary and the peculiar textual structure of patent documents. In this paper we analyze patents along the orthogonal dimensions of topic and textual structure. We view different patent classes and different patent text sections such as title, abstract, and claims, as separate translation tasks, and investigate the influence of such tasks on machine translation performance. We study multitask learning techniques that exploit commonalities between tasks by mixtures of translation models or by multi-task metaparameter tuning. We find small but significant gains over task-specific training by techniques that model commonalities through shared parameters. A by-product of our work is a parallel patent corpus of 23 million German-English sentence pairs. 1

Citations