文章详情

题名 · 作者 · 摘要 · 关键词 · DOI · 参考文献

中文文本拼写纠错研究综述

沈友志1，程春雷1，2，句泽东1，龚著凡1

（1.江西中医药大学 计算机学院，江西 南昌 330004；2.江西省中医人工智能重点研究室，江西 南昌 330004）

栏目：计算机技术 · 第 2025 年第 8 期 · 页码：138-145 · 2096-4706（2025）08-0138-08 · 中图分类号：TP391.1；TP301.1 · 文献标识码：A · DOI：10.19850/j.cnki.2096-4706.2025.08.025

摘要

中文文本拼写纠错（Chinese Spelling Correction, CSC）在自然语言处理（NLP）领域中是一个关键的基础任务，可为其下游任务和研究提供支持。CSC任务领域的研究持续发展，主要分为基于N-Gram语言模型、深度学习和大语言模型的纠错方法。首先对N-Gram语言模型的特性和在中文拼写纠错中的应用进行了分析，揭示了其在捕捉上下文信息方面的优势。其次基于深度学习的方法通过深度神经网络提高了纠错的准确性，并被广泛应用于中文文本处理中。同时，大语言模型的兴起为拼写纠错提供了新的思路，展示了其在处理复杂语言现象中的巨大潜力。该综述对中文文本拼写纠错领域的研究现状进行了详细的阐述，为从事相关研究人员提供参考。

关键词

中文文本；拼写纠错；N-Gram语言模型；深度学习；大语言模型

参考文献

  1. [1] KONDRAK G. N-Gram Similarity and Distance [C]//String Processing and Information Retrieval (SPIRE 2005).Buenos Aires：Springer Nature，2005：115-126.
  2. [2] XIE W J，HUANG P J，ZHANG X R，et al. Chinese Spelling Check System Based on N-Gram Model [C]//Proceedings of the Eighth SIGHAN Workshop on Chinese Language Processing (SIGHAN-8).Beijing：ACL，2015：128-136.
  3. [3] HUANG Q，HUANG P J，ZHANG X R，et al. Chinese Spelling Check System Based on Tri-Gram Model [C]//Proceedings of the Third CIPS-SIGHAN Joint Conference on Chinese Language Processing.Wuhan：ACL，2014：173-178.
  4. [4] YU J J，LI Z H. Chinese Spelling Error Detection and Correction Based on Language Model, Pronunciation, and Shape [C]//Proceedings of the Third CIPS-SIGHAN Joint Conference on Chinese Language Processing.Wuhan：ACL，2014：220-223.
  5. [5] YEH J-F，LI S-F，WU M-R，et al. Chinese Word Spelling Correction Based on N-Gram Ranked Inverted Index List [C]//Proceedings of the Seventh SIGHAN Workshop on Chinese Language Processing.Nagoya：ACL，2013：43-48.
  6. [6] YU L-C，LEE L-H，TSENG Y-H，et al. Overview of SIGHAN 2014 Bake-off for Chinese Spelling Check [C]//Proceedings of the Third CIPS-SIGHAN Joint Conference on Chinese Language Processing.Wuhan：ACL，2014：126-132.
  7. [7] VASWANI A，SHAZEER N，PARMAR N，et al. Attention Is All You Need [C]//31st International Conference on Neural Information Processing Systems.Long Beach：Curran Associates Inc，2017：6000-6010.
  8. [8] SCHMIDHUBER J. Deep Learning in Neural Networks: An Overview [J].Neural Networks，2015，61：85-117.
  9. [9] WANG D M，SONG Y，LI J，et al. A Hybrid Approach to Automatic Corpus Generation for Chinese Spelling Check [C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.Brussels：ACL，2018：2517-2527.
  10. [10] ZHANG S H，HUANG H R，LIU J C，et al. Spelling Error Correction with Soft-Masked BERT [C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.Online：ACL，2020：882-890.
  11. [11] ZHU C X，YING Z Q，ZHANG B Y，et al. MDCSpell: A Multi-task Detector-corrector Framework for Chinese Spelling Correction [C]//Findings of the Association for Computational Linguistics.Dublin：ACL，2022：1244-1253.
  12. [12] LIU S L，SONG S K，YUE T C，et al. CRASpell: A Contextual Typo Robust Approach to Improve Chinese Spelling Correction [C]//Findings of the Association for Computational Linguistics.Dublin：ACL，2022：3008-3018.
  13. [13] WU H Q，ZHANG S H，ZHANG Y C，et al. Rethinking Masked Language Modeling for Chinese Spelling Correction [C]//Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics.Toronto：ACL，2023：10743-10756.
  14. [14] LIU L F，WU H Q，ZHAO H. Chinese Spelling Correction as Rephrasing Language Model [J/OL].arXiv:2308.08796 [cs.CL].[2024-10-02].https://arxiv.org/abs/2308.08796.
  15. [15] 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. Online：ACL，2020：871-881.
  16. [16] XU H-D，LI Z L，ZHOU Q Y，et al. Read, Listen, and See: Leveraging Multimodal Information Helps Chinese Spell Checking [J/OL].arXiv:2105.12306 [cs.CL].[2024-10-02].https://arxiv.org/abs/2105.12306.
  17. [17] LI J H，WANG Q，MAO Z D，et al. Improving Chinese Spelling Check by Character Pronunciation Prediction：The Effects of Adaptivity and Granularity [J/OL].arXiv:2210.10996 [cs.CL].[2024-10-04].https://arxiv.org/abs/2210.10996.
  18. [18] LIANG Z H，QUAN X J，WANG Q F. Disentangled Phonetic Representation for Chinese Spelling Correction [J/OL].arXiv:2305.14783 [cs.CL].[2024-10-05].https://arxiv.org/abs/2305.14783?context=cs.CL.
  19. [19] WU H Q，ZHANG S H，ZHANG Y C，et al. Rethinking Masked Language Modeling for Chinese Spelling Correction [J/OL].arXiv:2305.17721 [cs.CL].[2024-10-07].https://arxiv.org/abs/2305.17721?context=cs.
  20. [20] ZHAO W X，ZHOU K，LI J Y，et al. A Survey of Large Language Models [J/OL].arXiv:2303.18223 [cs.CL].[2024-09-20].https://arxiv.org/abs/2303.18223.
  21. [21] KAPLAN J，MCCANDLISH S，HENIGHAN T，et al. Scaling Laws for Neural Language Models [J/OL].arXiv:2001.08361 [cs.LG].[2024-09-20].https://arxiv.org/abs/2001.08361?file=2001.08361.
  22. [22] WU T Y，HE S Z，LIU J P，et al. A Brief Overview of ChatGPT: The History, Status Quo and Potential Future Development [J].IEEE/CAA Journal of Automatica Sinica，2023，10（5）：1122-1136.
  23. [23] ANIL R，DAI A M，FIRAT O，et al. PaLM 2 Technical Report [J/OL].arXiv:2305.10403 [cs.CL].[2024-09-10].https://arxiv.org/abs/2305.10403v3#.
  24. [24] LI K T，HU Y，HE L，et al. C-LLM: Learn to Check Chinese Spelling Errors Character by Character [J/OL].arXiv:2406.16536 [cs.CL].[2024-09-10].https://arxiv.org/abs/2406.16536.
  25. [25] LI Y H，HUANG H J，MA S R，et al. On the （in） Effectiveness of Large Language Models for Chinese Text Correction [J/OL].arXiv:2307.09007 [cs.CL].[2024-09-16].https://arxiv.org/abs/2307.09007?context=cs.CL.
  26. [26] DONG M，CHEN Y J，ZHANG M，et al. Rich Semantic Knowledge Enhanced Large Language Models for Few-shot Chinese Spell Checking [J/OL].arXiv:2403.08492 [cs.CL].[2024-09-16].https://arxiv.org/abs/2403.08492.
  27. [27] ZHOU H Q，LI Z H，ZHANG B，et al. A Simple yet Effective Training-free Prompt-free Approach to Chinese Spelling Correction Based on Large Language Models [J/OL].arXiv:2410.04027 [cs.CL].[2024-09-16].https://arxiv.org/abs/2410.04027?context=cs.CL.

基金项目

江西省自然科学基金资助项目（20224BAB206102）；江西省教育厅科学技术研究项目（GJJ2200923）；国家自然科学基金（82260988 ）；江西省卫生和计划生育委员会-科技计划项目（202211404）

作者简介

沈友志（1997—），男，汉族，江西九江人，硕士在读，研究方向：自然语言处理；通信作者：程春雷（1976—），男，汉族，江西南昌人，副教授，硕士生导师，博士，研究方向：机器学习、知识表示与学习、知识图谱；句泽东（1998—），男，汉族，山西朔州人，硕士在读，研究方向：自然语言处理；龚著凡（2000—），男，汉族，江西南昌人，硕士在读，研究方向：自然语言处理。