# Rewriter-Evaluator Architecture for Neural Machine Translation

arXiv: https://arxiv.org/abs/2012.05414 · HTML: https://mlsanity.com/paper/2012.05414/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Dec 2020.

## Built on by (most cited first, 1 of 1)

- [Towards Table-to-Text Generation with Pretrained Language Model: A Table Structure Understanding and Text Deliberating Approach](https://mlsanity.com/paper/2301.02071/) (2023)

## Abstract (arXiv)

Encoder-decoder has been widely used in neural machine translation (NMT). A few methods have been proposed to improve it with multiple passes of decoding. However, their full potential is limited by a lack of appropriate termination policies. To address this issue, we present a novel architecture, Rewriter-Evaluator. It consists of a rewriter and an evaluator. Translating a source sentence involves multiple passes. At every pass, the rewriter produces a new translation to improve the past translation and the evaluator estimates the translation quality to decide whether to terminate the rewriting process. We also propose prioritized gradient descent (PGD) that facilitates training the rewriter and the evaluator jointly. Though incurring multiple passes of decoding, Rewriter-Evaluator with the proposed PGD method can be trained with a similar time to that of training encoder-decoder models. We apply the proposed architecture to improve the general NMT models (e.g., Transformer). We conduct extensive experiments on two translation tasks, Chinese-English and English-German, and show that the proposed architecture notably improves the performances of NMT models and significantly outperforms previous baselines.