# A Simple and Effective Approach to Automatic Post-Editing with Transfer Learning

arXiv: https://arxiv.org/abs/1906.06253 · HTML: https://mlsanity.com/paper/1906.06253/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2019.

## Built on by (most cited first, 11 of 11)

- [Leveraging GPT-4 for Automatic Translation Post-Editing](https://mlsanity.com/paper/2305.14878/) (2023)
- [Bridging the Gap: A Survey on Integrating (Human) Feedback for Natural Language Generation](https://mlsanity.com/paper/2305.00955/) (2023)
- [TEaR: Improving LLM-based Machine Translation with Systematic Self-Refinement](https://mlsanity.com/paper/2402.16379/) (2024)
- [A New Approach to Overgenerating and Scoring Abstractive Summaries](https://mlsanity.com/paper/2104.01726/) (2021)
- [Ladder: A Model-Agnostic Framework Boosting LLM-based Machine Translation to the Next Level](https://mlsanity.com/paper/2406.15741/) (2024)
- [Transfer Learning for Sequence Generation: from Single-source to Multi-source](https://mlsanity.com/paper/2105.14809/) (2021)
- [Automatic Correction of Human Translations](https://mlsanity.com/paper/2206.08593/) (2022)
- [PePe: Personalized Post-editing Model utilizing User-generated Post-edits](https://mlsanity.com/paper/2209.10139/) (2022)
- [Automatic Post-Editing for Vietnamese](https://mlsanity.com/paper/2104.12128/) (2021)
- [An Empirical Study of Automatic Post-Editing](https://mlsanity.com/paper/2209.07759/) (2022)
- [LangMark: A Multilingual Dataset for Automatic Post-Editing](https://mlsanity.com/paper/2511.17153/) (2025)

## Abstract (arXiv)

Automatic post-editing (APE) seeks to automatically refine the output of a black-box machine translation (MT) system through human post-edits. APE systems are usually trained by complementing human post-edited data with large, artificial data generated through back-translations, a time-consuming process often no easier than training an MT system from scratch. In this paper, we propose an alternative where we fine-tune pre-trained BERT models on both the encoder and decoder of an APE system, exploring several parameter sharing strategies. By only training on a dataset of 23K sentences for 3 hours on a single GPU, we obtain results that are competitive with systems that were trained on 5M artificial sentences. When we add this artificial data, our method obtains state-of-the-art results.