Review of Research Progress in Chinese Grammar Error Correction Technology

doi:10.3778/j.issn.1002-8331.2412-0070

Abstract

Abstract: The Chinese grammar error correction (CGEC) task aims to automatically correct Chinese sentences containing grammatical errors using natural language processing technology. This task is committed to improving the accuracy and readability of texts, enhancing the effect of information transmission, and showing its indispensable importance in many fields such as news releases, book publishing, voice input, and medical record quality control. This paper first reviews the development of Chinese grammar error correction technology, introduces the commonly used evaluation indicators and public datasets, and analyzes the main challenges faced by Chinese grammar error correction at present, including the lack of training library corpus and excessive correction. In view of these challenges, the paper systematically reviews the traditional error correction methods. Then, the research and application of sequence model method in the field of syntax error correction are further discussed, and they are mainly divided into two categories： sequence-to-sequence (Seq2Seq)-based error correction model and sequence-to-edit (Seq2Edit)-based error correction model. Finally, a new path of syntax error correction based on large language model is systematically sorted out. Based on the perspective of problem solving, this paper analyzes and summarizes various error correction models in detail, and finally looks forward to the challenges and future development directions of Chinese grammar error correction.

Key words: grammar correction, deep learning, large language models(LLMs), natural language processing technology

摘要： 中文语法错误纠正任务（Chinese grammar error correction，CGEC）旨在利用自然语言处理技术自动纠正含有语法错误的中文句子。该任务致力于提升文本的准确性与可读性，增强信息传递的效果，在新闻发布、书刊出版、语音输入、病历质控等多个领域展现出其不可或缺的重要性。回顾了中文语法纠错技术的发展脉络，介绍了常用的评价指标及公开数据集，并剖析当前中文语法纠错面临的主要挑战，包括训练库语料匮乏、过度纠正等问题。针对这些挑战，系统梳理传统的纠错方法；进而深入探讨序列模型方法在语法纠错领域的研究应用，将其主要划分为基于序列到序列（Seq2Seq）的纠错模型和基于序列到编辑（Seq2Edit）的纠错模型两大类别；系统地梳理了基于大语言模型的语法纠错新路径。立足问题解决视角，对各类纠错模型展开详尽分析与总结，最后展望了中文语法纠错面临的挑战与未来的发展方向。

关键词: 语法纠错, 深度学习, 大语言模型（LLMs）, 自然语言处理技术

JU Zedong, CHENG Chunlei, YE Qing, PENG Lin, GONG Zhufan. Review of Research Progress in Chinese Grammar Error Correction Technology[J]. Computer Engineering and Applications, 2025, 61(20): 36-53.

句泽东, 程春雷, 叶青, 彭琳, 龚著凡. 中文语法纠错技术的研究进展综述[J]. 计算机工程与应用, 2025, 61(20): 36-53.

References

[1] 李云汉, 施运梅, 李宁, 等. 中文文本自动校对综述[J]. 中文信息学报, 2022, 36(9): 1-18.
LI Y H, SHI Y M, LI N, et al. A survey of automatic error correction of Chinese text[J]. Journal of Chinese Information Processing, 2022, 36(9): 1-18.
[2] SHERSTINSKY A. Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network[J]. Physica D: Nonlinear Phenomena, 2020, 404: 132306.
[3] LEE C, HWANG Y G, OH H J, et al. Fine-grained named entity recognition using conditional random fields for question answering[C]//Proceedings of the 3rd Asia Information Retri-eval Technology SymposOum. Berlin: Heidelberg, 2006: 581-587.
[4] BORONG H, XUDONG L. Modern Chinese: updated five editions[M]. Beijing: Higher Education Press, 2011.
[5] YU L C, LEE L H, CHANG L P. Overview of grammatical error diagnosis for learning Chinese as a foreign language[C]//Proceedings of the 22nd International Conference on Computers in Education, 2014: 42-47.
[6] LEE L H, YU L C, CHANG L P. Overview of the NLP-TEA 2015 shared task for Chinese grammatical error diagnosis[C]//Proceedings of the 2nd Workshop on Natural Language Processing Techniques for Educational Applications. Stroudsburg: ACL, 2015: 1-6.
[7] LEE L H, YU L C, CHANG L P. Overview of the NLP-TEA 2016 shared task for Chinese grammatical error diagnosis[C]//Proceedings of the 2nd Workshop on Natural Language Processing Techniques for Educational Applications. Stroudsburg: ACL, 2015: 1-6.
[8] RAO G, ZHANG B, XUN E, et al. IJCNLP-2017 task 1: Chinese grammatical error diagnosis[C]//Proceedings of the International Joint Conference on Natural Language Processing, 2017: 1-8.
[9] RAO G Q, GONG Q, ZHANG B L, et al. Overview of NLPTEA-2018 share task Chinese grammatical error diagnosis[C]//Proceedings of the 5th Workshop on Natural Language Processing Techniques for Educational Applications. Stroudsburg: ACL, 2018: 42-51.
[10] RAO G Q, YANG E H, ZHANG B L. Overview of NLPTEA-2020 shared task for Chinese grammatical error diagnosis[C]//Proceedings of the 6th Workshop on Natural Language Processing Techniques for Educational Applications. Stroudsburg: ACL, 2020: 25-35.
[11] ZHAO Y Y, JIANG N, SUN W W, et al. Overview of the NLPCC 2018 shared task: grammatical error correction[C]//Proceedings of the Natural Language Processing and Chinese Computing. Cham: Springer International Publishing, 2018: 439-445.
[12] ZHANG Y, LI Z, BAO Z, et al. MuCGEC: a multi-reference multi-source evaluation dataset for Chinese grammatical error correction[J]. arXiv:2204.10994, 2022.
[13] XIN Y, ZHAO H, WANG Y Z, et al. An improved graph model for Chinese spell checking[C]//Proceedings of the 3rd CIPS-SIGHAN Joint Conference on Chinese Language Processing. Stroudsburg: ACL, 2014: 157-166.
[14] 刘亮亮, 曹存根. 中文“非多字词错误”自动校对方法研究[J]. 计算机科学, 2016, 43(10): 200-205.
LIU L L, CAO C G. Study of automatic proofreading method for non-multi-character word error in Chinese text[J]. Computer Science, 2016, 43(10): 200-205.
[15] LEE L H, CHANG L P, LEE K C, et al. Linguistic rules based Chinese error detection for second language learning[C]//Proceedings of the 21st International Conference on Computers in Education, 2013: 27-29.
[16] LEE L H, YU L C, LEE K C, et al. A sentence judgment system for grammatical error detection[C]//Proceedings of the 25th International Conference on Computational Linguistics: System Demonstrations, 2014: 67-70.
[17] CHANG T H, SUNG Y T, HONG J F, et al. KNGED: a tool for grammatical error diagnosis of Chinese sentences[C]//Proceedings of the 22nd International Conference on Computers in Education, 2014.
[18] WU C H, LIU C H, HARRIS M, et al. Sentence correction incorporating relative position and parse template language models[J]. IEEE Transactions on Audio, Speech, and Language Processing, 2010, 18(6): 1170-1181.
[19] FU K, HUANG J, DUAN Y T. Youdao’s winning solution to the NLPCC-2018 task 2 challenge: a neural machine translation approach to Chinese grammatical error correction[C]//Proceedings of the Natural Language Processing and Chinese Computing. Cham: Springer International Publishing, 2018: 341-350.
[20] CHEN P H, LIN C J, SCH?LKOPF B. A tutorial on ν-support vector machines[J]. Applied Stochastic Models in Business and Industry, 2005, 21(2): 111-136.
[21] LIU X, CHENG K, LUO Y, et al. A hybrid Chinese spelling correction using language model and statistical machine translation with reranking[C]//Proceedings of the 7th SIGHAN Workshop on Chinese Language Processing, 2013: 54-58.
[22] CHENG S M, YU C H, CHEN H H. Chinese word ordering errors detection and correction for non-native Chinese language learners[C]//Proceedings of the 25th International Conference on Computational Linguistics, 2014: 279-289.
[23] YU C H, CHEN H H. Detecting word ordering errors in Chinese sentences for learning Chinese as a foreign language[C]//Proceedings of the International Conference on Computational Linguistics, 2016: 3003-3018.
[24] CHANG R Y, WU C H, PRASETYO P K. Error diagnosis of Chinese sentences using inductive learning algorithm and decomposition-based testing mechanism[J]. ACM Transa-ctions on Asian Language Information Processing, 2012, 11(1): 1-24.
[25] XIONG J H, ZHANG Q, HOU J P, et al. Extended HMM and ranking models for Chinese spelling correction[C]//Proceedings of the 3rd CIPS-SIGHAN Joint Conference on Chinese Language Processing. Stroudsburg: ACL, 2014: 133-138.
[26] REN H K, YANG L E, XUN E D. A sequence to sequence learning for Chinese grammatical error correction[C]//Proceedings of the Natural Language Processing and Chinese Computing. Cham: Springer International Publishing, 2018: 401-410.
[27] ZHOU J P, LI C, LIU H Y, et al. Chinese grammatical error correction using statistical and neural models[C]//Proceedings of the Natural Language Processing and Chinese Computing. Cham: Springer International Publishing, 2018: 117-128.
[28] QIU Z Q, QU Y L. A two-stage model for Chinese grammatical error correction[J]. IEEE Access, 2019, 7: 146772-146777.
[29] WANG D M, TAY Y, ZHONG L. Confusionset-guided pointer networks for Chinese spelling check[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2019: 5780-5785.
[30] ZHAO Z W, WANG H F. MaskGEC: improving neural grammatical error correction via dynamic masking[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 1226-1233.
[31] CHENG X Y, XU W D, CHEN K L, et al. SpellGCN: incorporating phonological and visual similarities into language models for Chinese spelling check[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 871-881.
[32] 王辰成, 杨麟儿, 王莹莹, 等. 基于Transformer增强架构的中文语法纠错方法[J]. 中文信息学报, 2020, 34(6): 106-114.
WANG C C, YANG L E, WANG Y Y, et al. Chinese grammatical error correction method based on Transformer enha-nced architecture[J]. Journal of Chinese Information Processing, 2020, 34(6): 106-114.
[33] XIE H, CHEN Z, CHENG J, et al. Chinese grammar error detection based on data enhancement and multi-task feature learning[C]//Proceedings of the 19th Chinese National Conference on Computational Linguistics, 2020: 761-770.
[34] TANG Z, JI Y, ZHAO Y, et al. Chinese grammatical error correction enhanced by data augmentation from word and character levels[C]//Proceedings of the 20th Chinese National Conference on Computational Linguistics, 2021: 813-824.
[35] 李嘉诚, 沈嘉钰, 龚晨, 等. 基于指针网络融入混淆集知识的中文语法纠错[J]. 中文信息学报, 2022, 36(4): 29-38.
LI J C, SHEN J Y, GONG C, et al. Incorporating confusion set knowledge with pointer network for Chinese grammatical error correction[J]. Journal of Chinese Information Proces-sing, 2022, 36(4): 29-38.
[36] 孙邱杰, 梁景贵, 李思. 基于BART噪声器的中文语法纠错模型[J]. 计算机应用, 2022, 42(3): 860-866.
SUN Q J, LIANG J G, LI S. Chinese grammatical error correction model based on bidirectional and auto-regressive transformers noiser[J]. Journal of Computer Applications, 2022, 42(3): 860-866.
[37] 邓倩. 基于知识图谱的中文语法纠错研究[D]. 武汉: 华中师范大学, 2023.
DENG Q. Chinese grammatical error correction based on knowledge graph[D]. Wuhan: Central China Normal University, 2023.
[38] SU C, ZHAO X, QIAO X, et al. HWCGEC: HW-TSC’s 2023 submission for the NLPCC2023’s Chinese grammatical error correction task[C]//Proceedings of the CCF International Conference on Natural Language Processing and Chinese Computing. Cham: Springer Nature Switzerland, 2023: 59-68.
[39] ZHAO W, WANG L, SHEN K, et al. Improving grammatical error correction via pre-training a copy-augmented architecture with unlabeled data[J]. arXiv:1903.00138, 2019.
[40] WANG S L, WANG B X, GONG J F, et al. Combining ResNet and transformer for Chinese grammatical error diagnosis[C]//Proceedings of the 6th Workshop on Natural Language Processing Techniques for Educational Applications. Stroudsburg: ACL, 2020: 36-43.
[41] LUO Y K, BAO Z Y, LI C, et al. Chinese grammatical error diagnosis with graph convolution network and multi-task learning[C]//Proceedings of the 6th Workshop on Natural Language Processing Techniques for Educational Applications. Stroudsburg: ACL, 2020: 44-48.
[42] ZHANG Y, GOYAL R. SynGEC: syntax-enhanced grammatical error correction with a tailored GEC-oriented parser[C]//Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2022.
[43] ZHENG B, CHE W, GUO J, et al. Chinese grammatical error diagnosis with long short-term memory networks[C]//Proceedings of the 3rd Workshop on Natural Language Proces-sing Techniques for Educational Applications, 2016: 49-56.
[44] LIAO Q, WANG J, YANG J, et al. YNU-HPCC at IJCNLP-2017 task 1: Chinese grammatical error diagnosis using a bi-directional LSTM-CRF model[C]//Proceedings of the International Joint Conference on Natural Language Processing, 2017: 73-77.
[45] YANG Y, XIE P, TAO J, et al. Alibaba at IJCNLP-2017 task 1: embedding grammatical features into LSTMs for Chinese grammatical error diagnosis task[C]//Proceedings of the International Joint Conference on Natural Language Processing, 2017: 41-46.
[46] LIU Y J, ZAN H Y, ZHONG M J, et al. Detecting simultaneously Chinese grammar errors based on a BiLSTM-CRF model[C]//Proceedings of the 5th Workshop on Natural Language Processing Techniques for Educational Applications. Stroudsburg: ACL, 2018: 188-193.