# Grammatical Error Correction with Neural Reinforcement Learning

arXiv: https://arxiv.org/abs/1707.00299 · HTML: https://mlsanity.com/paper/1707.00299/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Jul 2017.

## Built on by (most cited first, 7 of 7)

- [Transparent Human Evaluation for Image Captioning](https://mlsanity.com/paper/2111.08940/) (2021)
- [Defending Pre-trained Language Models from Adversarial Word Substitutions Without Performance Sacrifice](https://mlsanity.com/paper/2105.14553/) (2021)
- [How do humans perceive adversarial text? A reality check on the validity and naturalness of word-based adversarial attacks](https://mlsanity.com/paper/2305.15587/) (2023)
- [gec-metrics: A Unified Library for Grammatical Error Correction Evaluation](https://mlsanity.com/paper/2505.19388/) (2025)
- [Residue-Based Natural Language Adversarial Attack Detection](https://mlsanity.com/paper/2204.10192/) (2022)
- [TREATED:Towards Universal Defense against Textual Adversarial Attacks](https://mlsanity.com/paper/2109.06176/) (2021)
- [Sentiment Perception Adversarial Attacks on Neural Machine Translation Systems](https://mlsanity.com/paper/2305.01437/) (2023)

## Abstract (arXiv)

We propose a neural encoder-decoder model with reinforcement learning (NRL) for grammatical error correction (GEC). Unlike conventional maximum likelihood estimation (MLE), the model directly optimizes towards an objective that considers a sentence-level, task-specific evaluation metric, avoiding the exposure bias issue in MLE. We demonstrate that NRL outperforms MLE both in human and automated evaluation metrics, achieving the state-of-the-art on a fluency-oriented GEC corpus.