Abstract
In this paper we present the annotation of COMPARA, currently the largest parallel corpora which includes Portuguese. We describe the motivation, give a glimpse of the results so far, and the way the corpus is being annotated, as well as mention some studies based on it.