Can Neural Machine Translation be Improved with User Feedback?

TL;DR

This paper demonstrates that real-world user feedback, especially implicit feedback from e-commerce tasks, can be effectively used to improve neural machine translation models, unlike explicit ratings which are less reliable.

Contribution

It is the first to apply real-world user feedback for offline bandit learning in NMT, highlighting the effectiveness of implicit feedback over explicit ratings.

Findings

Implicit task-based feedback improves translation quality.

Explicit five-star ratings are unreliable for model improvement.

Real logged feedback enables practical NMT enhancements.

Abstract

We present the first real-world application of methods for improving neural machine translation (NMT) with human reinforcement, based on explicit and implicit user feedback collected on the eBay e-commerce platform. Previous work has been confined to simulation experiments, whereas in this paper we work with real logged feedback for offline bandit learning of NMT parameters. We conduct a thorough analysis of the available explicit user judgments---five-star ratings of translation quality---and show that they are not reliable enough to yield significant improvements in bandit learning. In contrast, we successfully utilize implicit task-based feedback collected in a cross-lingual search task to improve task-specific and machine translation quality metrics.