Achieving Human Parity on Automatic Chinese to English News Translation
This addresses the challenge of cross-language communication for users of online translation systems by demonstrating a significant advancement in translation quality.
The paper tackled the problem of defining and measuring human parity in machine translation, and found that Microsoft's neural machine translation system achieved state-of-the-art results, reaching human parity with professional translations on the WMT 2017 Chinese to English news task.
Machine translation has made rapid advances in recent years. Millions of people are using it today in online translation systems and mobile applications in order to communicate across language barriers. The question naturally arises whether such systems can approach or achieve parity with human translations. In this paper, we first address the problem of how to define and accurately measure human parity in translation. We then describe Microsoft's machine translation system and measure the quality of its translations on the widely used WMT 2017 news translation task from Chinese to English. We find that our latest neural machine translation system has reached a new state-of-the-art, and that the translation quality is at human parity when compared to professional human translations. We also find that it significantly exceeds the quality of crowd-sourced non-professional translations.