Salvato in:
Dettagli Bibliografici
Autori principali: Zhong, Yufeng, Zeng, Zhixiong, Chen, Lei, Yang, Longrong, Zheng, Liming, Huang, Jing, Yang, Siqi, Ma, Lin
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2508.00311
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908475171274752
author Zhong, Yufeng
Zeng, Zhixiong
Chen, Lei
Yang, Longrong
Zheng, Liming
Huang, Jing
Yang, Siqi
Ma, Lin
author_facet Zhong, Yufeng
Zeng, Zhixiong
Chen, Lei
Yang, Longrong
Zheng, Liming
Huang, Jing
Yang, Siqi
Ma, Lin
contents Optical Character Recognition (OCR) for mathematical formula is essential for the intelligent analysis of scientific literature. However, both task-specific and general vision-language models often struggle to handle the structural diversity, complexity, and real-world variability inherent in mathematical content. In this work, we present DocTron-Formula, a unified framework built upon general vision-language models, thereby eliminating the need for specialized architectures. Furthermore, we introduce CSFormula, a large-scale and challenging dataset that encompasses multidisciplinary and structurally complex formulas at the line, paragraph, and page levels. Through straightforward supervised fine-tuning, our approach achieves state-of-the-art performance across a variety of styles, scientific domains, and complex layouts. Experimental results demonstrate that our method not only surpasses specialized models in terms of accuracy and robustness, but also establishes a new paradigm for the automated understanding of complex scientific documents.
format Preprint
id arxiv_https___arxiv_org_abs_2508_00311
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DocTron-Formula: Generalized Formula Recognition in Complex and Structured Scenarios
Zhong, Yufeng
Zeng, Zhixiong
Chen, Lei
Yang, Longrong
Zheng, Liming
Huang, Jing
Yang, Siqi
Ma, Lin
Computer Vision and Pattern Recognition
Optical Character Recognition (OCR) for mathematical formula is essential for the intelligent analysis of scientific literature. However, both task-specific and general vision-language models often struggle to handle the structural diversity, complexity, and real-world variability inherent in mathematical content. In this work, we present DocTron-Formula, a unified framework built upon general vision-language models, thereby eliminating the need for specialized architectures. Furthermore, we introduce CSFormula, a large-scale and challenging dataset that encompasses multidisciplinary and structurally complex formulas at the line, paragraph, and page levels. Through straightforward supervised fine-tuning, our approach achieves state-of-the-art performance across a variety of styles, scientific domains, and complex layouts. Experimental results demonstrate that our method not only surpasses specialized models in terms of accuracy and robustness, but also establishes a new paradigm for the automated understanding of complex scientific documents.
title DocTron-Formula: Generalized Formula Recognition in Complex and Structured Scenarios
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.00311