arXiv 2510.145282025-11-25cs.CV

PaddleOCR-VL: 通过0.9B超紧凑视觉-语言模型提升多语言文档解析

PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model

PaddlePaddle Team, Baidu Inc.（百度公司）

Cheng Cui, Ting Sun, Suyin Liang, Tingquan Gao, Zelun Zhang, Jiaxuan Liu, Xueqing Wang, Changda Zhou, Hongen Liu, Manhui Lin, Yue Zhang, Yubo Zhang, Handong Zhe… 展开作者

Cheng Cui, Ting Sun, Suyin Liang, Tingquan Gao, Zelun Zhang, Jiaxuan Liu, Xueqing Wang, Changda Zhou, Hongen Liu, Manhui Lin, Yue Zhang, Yubo Zhang, Handong Zheng, Jing Zhang, Jun Zhang, Yi Liu, Dianhai Yu, Yanjun Ma

备注：: Github Repo: https://github.com/PaddlePaddle/PaddleOCR

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

PaddleOCR-VL通过0.9B超紧凑视觉-语言模型实现多语言文档解析的高效准确识别

摘要：

本文提出PaddleOCR-VL，一种最先进的且资源高效的文档解析模型。其核心组件是PaddleOCR-VL-0.9B，一个紧凑而强大的视觉-语言模型（VLM），它集成了NaViT风格的动态分辨率视觉编码器和ERNIE-4.5-0.3B语言模型，以实现准确的元素识别。该创新模型高效支持109种语言，并在识别复杂元素（如文本、表格、公式和图表）方面表现出色，同时保持极低的资源消耗。通过在广泛使用的公共基准和内部基准上的全面评估，PaddleOCR-VL在页面级文档解析和元素级识别方面均达到最先进的性能。它显著优于现有解决方案，展现出对顶级VLMs的强大竞争力，并提供快速的推理速度。这些优势使其在实际应用场景中具有高度的适用性。代码可在https://github.com/PaddlePaddle/PaddleOCR上获取。

英文摘要：

In this report, we propose PaddleOCR-VL, a SOTA and resource-efficient model tailored for document parsing. Its core component is PaddleOCR-VL-0.9B, a compact yet powerful vision-language model (VLM) that integrates a NaViT-style dynamic resolution visual encoder with the ERNIE-4.5-0.3B language model to enable accurate element recognition. This innovative model efficiently supports 109 languages and excels in recognizing complex elements (e.g., text, tables, formulas, and charts), while maintaining minimal resource consumption. Through comprehensive evaluations on widely used public benchmarks and in-house benchmarks, PaddleOCR-VL achieves SOTA performance in both page-level document parsing and element-level recognition. It significantly outperforms existing solutions, exhibits strong competitiveness against top-tier VLMs, and delivers fast inference speeds. These strengths make it highly suitable for practical deployment in real-world scenarios. Code is available at https://github.com/PaddlePaddle/PaddleOCR .

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。