搜索文章:
期刊:
主题:

期刊目次

2026年·卷4·期  1
2025年·卷3·期  1 / 2
2024年·卷2·期  1 / 2
2023年·卷1·期  1
综述 · Review

AI 赋能 Tex 公式智能生成技术应用

Application of AI-enabled Intelligent TeX Formula Generation Technology


作者:赵梦卓*
 西安工业大学 陕西 西安
*通信作者:赵梦卓;单位:西安工业大学 陕西 西安
AI应用研究, 2026, 4(1), 1-8; https://doi.org/10.58244/aiar.263949
提交日期 : 2026年01月20日 丨 录用日期 : 2026年03月17日 丨 出版日期 : 2026年06月20日
课题资助:自筹经费,无利益冲突需要说明
引用本文
摘 要:
在学术写作、科研报告、教学课件制作等场景中,Tex公式凭借其规范、美观的排版优势,成为数学、物理、计算机、化学等理工科及交叉学科不可或缺的核心表达工具。但传统Tex公式编写需熟练掌握复杂的语法命令与排版规则,不仅耗时费力,且极易出现符号错误、格式混乱等问题,严重制约了内容创作的效率与专业性。随着人工智能技术的快速迭代,尤其是豆包、DeepSeek等大语言模型与计算机视觉技术的深度融合,AI生成Tex公式技术应运而生,生成的代码可直接在WPS公式功能中复制粘贴使用,无需额外调试,大幅降低了Tex公式的使用门槛,让非专业使用者也能快速编辑出规范、美观的公式。本文从应用背景出发,详细介绍AI生成Tex公式的核心定义、特点与适用场景,拆解其技术框架的核心模块及工作原理,结合豆包、DeepSeek生成Tex代码并在WPS中粘贴使用的实操案例,补充多个场景图片及详细描写,分步讲解使用方法,最后对该技术的应用价值、现存不足及未来发展方向进行全面总结评价,为科研工作者、在校学生、教师及相关从业者提供实用、易懂的科普参考,助力该技术的广泛普及与合理应用。
关键词:AI生成Tex公式;人工智能;Tex公式;公式编辑;科普;学术排版;豆包;DeepSeek;WPS
 
Abstract:
In academic writing, research report compilation, courseware design and other related work, TeX formulas are widely adopted across science, engineering and interdisciplinary fields including mathematics, physics, computer science and chemistry. Renowned for their standardized layout and neat visual presentation, they have become an indispensable tool for professional content expression. However, traditional TeX formula writing requires proficient command of intricate syntax and formatting rules. This old approach is time-consuming and labor-intensive, and often leads to symbol errors and messy formatting, greatly undermining writing efficiency and content professionalism.With the rapid advancement of artificial intelligence, large language models and computer vision technologies have achieved in-depth integration, giving rise to the AI-based TeX formula generation technology. The generated codes can be directly copied and pasted into the formula function of WPS for immediate use with no extra revision, significantly lowering the barriers to TeX application. It enables ordinary users without professional expertise to create well-formatted and visually appealing formulas effortlessly. Starting with the practical application background, this paper elaborates on the definition, features and applicable scenarios of AI-generated TeX formulas, and analyzes the core components and operating mechanisms of its technical framework. With practical cases of generating TeX codes via mainstream platforms and direct application in WPS, supplemented by detailed illustrations and step-by-step guidance, this paper demonstrates the whole operation process. In the end, it comprehensively discusses the practical value, existing limitations and future development prospects of the technology. This study aims to offer accessible and practical references for researchers, students, teachers and relevant practitioners, and promote the popularization and rational application of AI TeX generation technology.
Keywords: AI-generated TeX formulas; Artificial intelligence; TeX formula; Formula editing; Popular science; Academic typesetting; Doubao; DeepSeek; WPS
 
--
正文内容 / Content:
可下载并阅读全文PDF,请按照本文版权许可使用。
Download the full text PDF for viewing and using it according to the license of this paper.

参考文献 / References: 
  1. Sundararaj J, Vyas A, Gonzalez-Maldonado B. Automated LaTeX Code Generation from Handwritten Mathematical Expressions[R]. arXiv:2412.03853, 2024.
  2. Shojaee P, Liao Y, Bach S, et al. LLM-SR: Scientific Equation Discovery via Programming with Large Language Models[C]//ICLR 2025. OpenReview, 2025.
  3. Han X, Ning C, Zhong J, et al. Discovering Mathematical Equations with Diffusion Language Model[R]. arXiv:2509.13136, 2025. 
  4. Song Z, Zhou Q, Ren C, et al. LLM-Feynman: Leveraging Large Language Models for Universal Scientific Formula and Theory Discovery[R]. arXiv:2503.06512, 2025.
  5.  Deng Y, Kanervisto A, Rush A M. Translating Mathematical Formula Images to LaTeX Sequences Using Deep Neural Networks with Sequence-level Training[R]. arXiv:1908.11415, 2019.
  6. Misra K, Saini R, Kumar A. PDF2LaTeX: A Deep Learning System to Convert Mathematical Documents from PDF to LaTeX[C]//Proceedings of the 4th ACM SIGSOFT International Workshop on Machine Learning Techniques for Software Quality Evaluation. ACM, 2021.
  7. 文墨共鸣模型处理LaTeX科学文档:从文本到排版公式的智能转换[EB/OL]. 2026-04-10.
  8. Qwen2.5-7B-Instruct LaTeX文档生成:自动化科研论文写作[EB/OL], 2026-03-21.
  9. 基于MiniCPM-V-2_6的智能文档处理:LaTeX自动生成[EB/OL].2026-02-14.
  10. LaTeX排版自动化:HunyuanOCR识别手写公式转LaTeX代码实验[EB/OL], 2026-01-02.  
Submit Your Manuscript Now