MonkeyOCR v1.5 Technical Report: Unlocking Robust Document Parsing for Complex Patterns
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911267722100736 |
|---|---|
| author | Zhang, Jiarui Liu, Yuliang Wu, Zijun Pang, Guosheng Ye, Zhili Zhong, Yupei Ma, Junteng Wei, Tao Xu, Haiyang Chen, Weikai Wang, Zeen Ji, Qiangjun Zhou, Fanxi Zhang, Qi Hu, Yuanrui Liu, Jiahao Li, Zhang Zhang, Ziyang Liu, Qiang Bai, Xiang |
| author_facet | Zhang, Jiarui Liu, Yuliang Wu, Zijun Pang, Guosheng Ye, Zhili Zhong, Yupei Ma, Junteng Wei, Tao Xu, Haiyang Chen, Weikai Wang, Zeen Ji, Qiangjun Zhou, Fanxi Zhang, Qi Hu, Yuanrui Liu, Jiahao Li, Zhang Zhang, Ziyang Liu, Qiang Bai, Xiang |
| contents | Document parsing is a core task in document intelligence, supporting applications such as information extraction, retrieval-augmented generation, and automated document analysis. However, real-world documents often feature complex layouts with multi-level tables, embedded images or formulas, and cross-page structures, which remain challenging for existing OCR systems. We introduce MonkeyOCR v1.5, a unified vision-language framework that enhances both layout understanding and content recognition through a two-stage pipeline. The first stage employs a large multimodal model to jointly predict layout and reading order, leveraging visual information to ensure sequential consistency. The second stage performs localized recognition of text, formulas, and tables within detected regions, maintaining high visual fidelity while reducing error propagation. To address complex table structures, we propose a visual consistency-based reinforcement learning scheme that evaluates recognition quality via render-and-compare alignment, improving structural accuracy without manual annotations. Additionally, two specialized modules, Image-Decoupled Table Parsing and Type-Guided Table Merging, are introduced to enable reliable parsing of tables containing embedded images and reconstruction of tables crossing pages or columns. Comprehensive experiments on OmniDocBench v1.5 demonstrate that MonkeyOCR v1.5 achieves state-of-the-art performance, outperforming PPOCR-VL and MinerU 2.5 while showing exceptional robustness in visually complex document scenarios. A trial link can be found at https://github.com/Yuliang-Liu/MonkeyOCR . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_10390 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MonkeyOCR v1.5 Technical Report: Unlocking Robust Document Parsing for Complex Patterns Zhang, Jiarui Liu, Yuliang Wu, Zijun Pang, Guosheng Ye, Zhili Zhong, Yupei Ma, Junteng Wei, Tao Xu, Haiyang Chen, Weikai Wang, Zeen Ji, Qiangjun Zhou, Fanxi Zhang, Qi Hu, Yuanrui Liu, Jiahao Li, Zhang Zhang, Ziyang Liu, Qiang Bai, Xiang Computer Vision and Pattern Recognition Artificial Intelligence Document parsing is a core task in document intelligence, supporting applications such as information extraction, retrieval-augmented generation, and automated document analysis. However, real-world documents often feature complex layouts with multi-level tables, embedded images or formulas, and cross-page structures, which remain challenging for existing OCR systems. We introduce MonkeyOCR v1.5, a unified vision-language framework that enhances both layout understanding and content recognition through a two-stage pipeline. The first stage employs a large multimodal model to jointly predict layout and reading order, leveraging visual information to ensure sequential consistency. The second stage performs localized recognition of text, formulas, and tables within detected regions, maintaining high visual fidelity while reducing error propagation. To address complex table structures, we propose a visual consistency-based reinforcement learning scheme that evaluates recognition quality via render-and-compare alignment, improving structural accuracy without manual annotations. Additionally, two specialized modules, Image-Decoupled Table Parsing and Type-Guided Table Merging, are introduced to enable reliable parsing of tables containing embedded images and reconstruction of tables crossing pages or columns. Comprehensive experiments on OmniDocBench v1.5 demonstrate that MonkeyOCR v1.5 achieves state-of-the-art performance, outperforming PPOCR-VL and MinerU 2.5 while showing exceptional robustness in visually complex document scenarios. A trial link can be found at https://github.com/Yuliang-Liu/MonkeyOCR . |
| title | MonkeyOCR v1.5 Technical Report: Unlocking Robust Document Parsing for Complex Patterns |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2511.10390 |