This work presents a novel approach to jointly tackling Automatic Post-Editing (APE) and Word-Level Quality Estimation (QE) using ensembles of specialized Neural Machine Translation (NMT) systems. Word-level features which have proven effective for QE are included as input factors, expanding the representation of the original source and the machine translation hypothesis, which are used to generate an automatically post-edited hypothesis. We train a suite of NMT models which use different input representations, but share the same output space. These models are then ensembled together, and tuned for both the APE and the QE task. We thus attempt to connect the state-of-the-art approaches to APE and QE within a single framework. Our models achieve state-of-the-art results in both tasks, with the only difference in the tuning step which learns weights for each component of the ensemble.

利用多个专门的神经机器翻译系统，将已被证明在词级质量估计中有效的特征作为输入因素，扩展原始源和机器翻译假设的表示，生成自动后编辑的假设，并在单个框架内连接自动后编辑和词级质量估计的最先进方法，进而通过调整实现两个任务的最优表现。

基于分解的神经机器翻译模型集成用于自动后编辑和质量评估