Papers › Unsupervised Grammatical Error Correction Rivaling Supervised Methods
Unsupervised Grammatical Error Correction Rivaling Supervised Methods
Hannan Cao, Liping Yuan, Yuchen Zhang, Hwee Tou Ng
State-of-the-art grammatical error correction (GEC) systems rely on parallel training data (ungrammatical sentences and their manually corrected counterparts), which are expensive to construct. In this paper, we employ the Break-It-Fix-It (BIFI) method to build an unsupervised GEC system. The BIFI framework generates parallel data from unlabeled text using a fixer to transform ungrammatical sentences into grammatical ones, and a critic to predict sentence grammaticality. We present an unsupervised approach to build the fixer and the critic, and an algorithm that allows them to iteratively improve each other. We evaluate our unsupervised GEC system on English and Chinese GEC. Empirical results show that our GEC system outperforms previous unsupervised GEC systems, and achieves performance comparable to supervised GEC systems without ensemble. Furthermore, when combined with labeled training data, our system achieves new state-of-the-art results on the CoNLL-2014 and NLPCC-2018 test sets.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Grammatical Error Correction | BEA-2019 (test) | Unsupervised GEC + cLang8 | F0.5 | 76.5 | #6 of 19 | Archive leaderboard | report |
| Grammatical Error Correction | CoNLL-2014 Shared Task | Unsupervised GEC + cLang8 | F0.5 | 69.6 | #5 of 23 | Archive leaderboard | report |
| Grammatical Error Correction | CoNLL-2014 Shared Task | Unsupervised GEC + cLang8 | Precision | 75.0 | #5 of 23 | Archive leaderboard | report |
| Grammatical Error Correction | CoNLL-2014 Shared Task | Unsupervised GEC + cLang8 | Recall | 53.8 | #5 of 23 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections