# APE at Scale and its Implications on MT Evaluation Biases

## TL;DR

This paper introduces an Automatic Post-Editing (APE) model trained on monolingual data to identify and correct biases in MT evaluation, revealing discrepancies between human judgments and BLEU scores.

## Contribution

It presents a novel APE approach trained on monolingual data that exposes biases in standard MT evaluation metrics and improves translation quality in multiple language tasks.

## Key findings

- Human-judged quality improves after APE correction.
- BLEU scores decrease on forward-translated test sets.
- Quality improvements of up to 2.5 BLEU points on WMT tasks.

## Abstract

In this work, we train an Automatic Post-Editing (APE) model and use it to reveal biases in standard Machine Translation (MT) evaluation procedures. The goal of our APE model is to correct typical errors introduced by the translation process, and convert the "translationese" output into natural text. Our APE model is trained entirely on monolingual data that has been round-trip translated through English, to mimic errors that are similar to the ones introduced by NMT. We apply our model to the output of existing NMT systems, and demonstrate that, while the human-judged quality improves in all cases, BLEU scores drop with forward-translated test sets. We verify these results for the WMT18 English to German, WMT15 English to French, and WMT16 English to Romanian tasks. Furthermore, we selectively apply our APE model on the output of the top submissions of the most recent WMT evaluation campaigns. We see quality improvements on all tasks of up to 2.5 BLEU points.