INTRODUCTION

摘要

The goal of quality estimation is to evaluate a translation system's quality without access to reference translations (Blatz et al., 2004; Specia et al., 2013). This has many potential usages: informing an end user about the reliability of translated content; deciding if a translation is ready for publishing or if it requires human post-editing; highlighting the words that need to be changed. Quality estimation systems are particularly appealing for crowd-sourced and professional translation services, due to their potential to dramatically reduce post-editing times and to save labor costs (Specia, 2011). The increasing interest in this problem from an industrial angle comes as no surprise (Turchi et al., 2014; de Souza et al., 2015; Martins et al., 2016, 2017; Kozlova et al., 2016). A related task is that of automatic post-editing (Simard et al. (2007), Junczys-Dowmunt and Grundkiewicz (2016)), which aims to automatically correct the output of machine translation. Recent work (Martins, 2017, Kim et al., 2017, Hokamp, 2017) has shown that the tasks of quality estimation and automatic post-editing benefit from being trained or stacked together.