Character Mapping and Ad-hoc Adaptation: Edinburghs IWSLT 2020 Open Domain Translation System

This paper describes the University of Edinburghs neural machine translation systems submitted to the IWSLT 2020 open domain Japanese$\leftrightarrow$Chinese translation task. On top of commonplace techniques like tokenisation and corpus cleaning, we explore character mapping and unsupervised decoding-time adaptation. Our techniques focus on leveraging the provided data, and we show the positive impact of each technique through the gradual improvement of BLEU.