# Leveraging What's Overfixed: Post-Correction via LLM Grammatical Error Overcorrection

arXiv: https://arxiv.org/abs/2509.20811 · HTML: https://mlsanity.com/paper/2509.20811/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 45 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Builds on (most cited first, 22 of 22)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [GECToR -- Grammatical Error Correction: Tag, Not Rewrite](https://mlsanity.com/paper/2005.12592/) (2020)
- [A Simple Recipe for Multilingual Grammatical Error Correction](https://mlsanity.com/paper/2106.03830/) (2021)
- [Encode, Tag, Realize: High-Precision Text Editing](https://mlsanity.com/paper/1909.01187/) (2019)
- [Is ChatGPT a Highly Fluent Grammatical Error Correction System? A Comprehensive Evaluation](https://mlsanity.com/paper/2304.01746/) (2023)
- [Parallel Iterative Edit Models for Local Sequence Transduction](https://mlsanity.com/paper/1910.02893/) (2019)
- [ChatGPT or Grammarly? Evaluating ChatGPT on Grammatical Error Correction Benchmark](https://mlsanity.com/paper/2303.13648/) (2023)
- [Seq2Edits: Sequence Transduction Using Span-level Edit Operations](https://mlsanity.com/paper/2009.11136/) (2020)
- [Approaching Neural Grammatical Error Correction as a Low-Resource Machine Translation Task](https://mlsanity.com/paper/1804.05940/) (2018)
- [Synthetic Data Generation for Grammatical Error Correction with Tagged Corruption Models](https://mlsanity.com/paper/2105.13318/) (2021)
- [Stronger Baselines for Grammatical Error Correction Using Pretrained Encoder-Decoder Model](https://mlsanity.com/paper/2005.11849/) (2020)
- [LM-Critic: Language Models for Unsupervised Grammatical Error Correction](https://mlsanity.com/paper/2109.06822/) (2021)
- [Neural Language Correction with Character-Based Attention](https://mlsanity.com/paper/1603.09727/) (2016)
- [Revisiting Grammatical Error Correction Evaluation and Beyond](https://mlsanity.com/paper/2211.01635/) (2022)
- [GPT-3.5 for Grammatical Error Correction](https://mlsanity.com/paper/2405.08469/) (2024)
- [A Nested Attention Neural Hybrid Model for Grammatical Error Correction](https://mlsanity.com/paper/1707.02026/) (2017)
- [Evaluating Prompting Strategies for Grammatical Error Correction Based on Language Proficiency](https://mlsanity.com/paper/2402.15930/) (2024)
- [Non-autoregressive Text Editing with Copy-aware Latent Alignments](https://mlsanity.com/paper/2310.07821/) (2023)
- [Key-Element-Informed sLLM Tuning for Document Summarization](https://mlsanity.com/paper/2406.04625/) (2024)
- [Alirector: Alignment-Enhanced Chinese Grammatical Error Corrector](https://mlsanity.com/paper/2402.04601/) (2024)
- [Efficient and Interpretable Grammatical Error Correction with Mixture of Experts](https://mlsanity.com/paper/2410.23507/) (2024)

## Built on by (most cited first, 2 of 2)

- [When VLMs 'Fix' Students: Identifying and Penalizing Over-Correction in the Evaluation of Multi-line Handwritten Math OCR](https://mlsanity.com/paper/2604.22774/) (2026)
- [Listen, Correct, and Feed Back: Spoken Pedagogical Feedback Generation](https://mlsanity.com/paper/2604.14177/) (2026)

## Abstract (arXiv)

Robust supervised fine-tuned small Language Models (sLMs) often show high reliability but tend to undercorrect. They achieve high precision at the cost of low recall. Conversely, Large Language Models (LLMs) often show the opposite tendency, making excessive overcorrection, leading to low precision. To effectively harness the strengths of LLMs to address the recall challenges in sLMs, we propose Post-Correction via Overcorrection (PoCO), a novel approach that strategically balances recall and precision. PoCO first intentionally triggers overcorrection via LLM to maximize recall by allowing comprehensive revisions, then applies a targeted post-correction step via fine-tuning smaller models to identify and refine erroneous outputs. We aim to harmonize both aspects by leveraging the generative power of LLMs while preserving the reliability of smaller supervised models. Our extensive experiments demonstrate that PoCO effectively balances GEC performance by increasing recall with competitive precision, ultimately improving the overall quality of grammatical error correction.