Quality expectations of machine translation

TL;DR

This paper discusses the current state of machine translation, focusing on quality evaluation methods, deployment practices, and how evolving quality standards impact its usefulness, especially for human translators.

Contribution

It provides an analysis of how MT quality assessment and deployment strategies need to adapt as translation quality improves and standards evolve.

Findings

MT is widely used for various applications by millions daily.

Current quality evaluation methods are insufficient as standards evolve.

The 'gold standard' for quality is no longer fixed, affecting deployment and acceptance.

Abstract

Machine Translation (MT) is being deployed for a range of use-cases by millions of people on a daily basis. There should, therefore, be no doubt as to the utility of MT. However, not everyone is convinced that MT can be useful, especially as a productivity enhancer for human translators. In this chapter, I address this issue, describing how MT is currently deployed, how its output is evaluated and how this could be enhanced, especially as MT quality itself improves. Central to these issues is the acceptance that there is no longer a single 'gold standard' measure of quality, such that the situation in which MT is deployed needs to be borne in mind, especially with respect to the expected 'shelf-life' of the translation itself.