# GitHub Typo Corpus: A Large-Scale Multilingual Dataset of Misspellings and Grammatical Errors

arXiv: https://arxiv.org/abs/1911.12893 · HTML: https://mlsanity.com/paper/1911.12893/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Nov 2019.

## Built on by (most cited first, 22 of 22)

- [Grammatical Error Correction: A Survey of the State of the Art](https://mlsanity.com/paper/2211.05166/) (2022)
- [Two-in-One: A Model Hijacking Attack Against Text Generation Models](https://mlsanity.com/paper/2305.07406/) (2023)
- [Neural Quality Estimation with Multiple Hypotheses for Grammatical Error Correction](https://mlsanity.com/paper/2105.04443/) (2021)
- [VarCLR: Variable Semantic Representation Pre-training via Contrastive Learning](https://mlsanity.com/paper/2112.02650/) (2021)
- [Correcting diacritics and typos with a ByT5 transformer model](https://mlsanity.com/paper/2201.13242/) (2022)
- [Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise?](https://mlsanity.com/paper/2306.15268/) (2023)
- [HydraOpt: Navigating the Efficiency-Performance Trade-off of Adapter Merging](https://mlsanity.com/paper/2507.17706/) (2025)
- [A Methodology for Generative Spelling Correction via Natural Spelling Errors Emulation across Multiple Domains and Languages](https://mlsanity.com/paper/2308.09435/) (2023)
- [K-Merge: Online Continual Merging of Adapters for On-device Large Language Models](https://mlsanity.com/paper/2510.13537/) (2025)
- [Data Augmentation for Intent Classification](https://mlsanity.com/paper/2206.05790/) (2022)
- [Misspellings in Natural Language Processing: A survey](https://mlsanity.com/paper/2501.16836/) (2025)
- [How You Ask Matters! Adaptive RAG Robustness to Query Variations](https://mlsanity.com/paper/2604.10745/) (2026)
- [READIN: A Chinese Multi-Task Benchmark with Realistic and Diverse Input Noises](https://mlsanity.com/paper/2302.07324/) (2023)
- [Multi-teacher Distillation for Multilingual Spelling Correction](https://mlsanity.com/paper/2311.11518/) (2023)
- [Clotho: Measuring Task-Specific Pre-Generation Test Adequacy for LLM Inputs](https://mlsanity.com/paper/2509.17314/) (2025)
- [Library Hallucinations in LLM-Generated Code: A Risk Analysis Grounded in Developer Queries](https://mlsanity.com/paper/2509.22202/) (2025)
- [Make Text Unlearnable: Exploiting Effective Patterns to Protect Personal Data](https://mlsanity.com/paper/2307.00456/) (2023)
- [Domain specificity and data efficiency in typo tolerant spell checkers: the case of search in online marketplaces](https://mlsanity.com/paper/2308.01976/) (2023)
- [Finite-context Indexing of Restricted Output Space for NLP Models Facing Noisy Input](https://mlsanity.com/paper/2310.14110/) (2023)
- [Data-driven Clustering and Merging of Adapters for On-device Large Language Models](https://mlsanity.com/paper/2601.17441/) (2026)
- [Low-Latency Spell Correction for Japanese Music Search Queries](https://mlsanity.com/paper/2609.04262/) (2026)
- [Evaluating the Robustness of Japanese LLMs to IME-Related and Typographical Errors](https://mlsanity.com/paper/2610.01241/) (2026)

## Abstract (arXiv)

The lack of large-scale datasets has been a major hindrance to the development of NLP tasks such as spelling correction and grammatical error correction (GEC). As a complementary new resource for these tasks, we present the GitHub Typo Corpus, a large-scale, multilingual dataset of misspellings and grammatical errors along with their corrections harvested from GitHub, a large and popular platform for hosting and sharing git repositories. The dataset, which we have made publicly available, contains more than 350k edits and 65M characters in more than 15 languages, making it the largest dataset of misspellings to date. We also describe our process for filtering true typo edits based on learned classifiers on a small annotated subset, and demonstrate that typo edits can be identified with F1 ~ 0.9 using a very simple classifier with only three features. The detailed analyses of the dataset show that existing spelling correctors merely achieve an F-measure of approx. 0.5, suggesting that the dataset serves as a new, rich source of spelling errors that complement existing datasets.