OvisOCR3 技術報告
OvisOCR3 Technical Report
OvisOCR3 是一款 0.8B 視覺語言模型,透過多教師知識蒸餾和強化學習,整合五類 OCR 相關任務。在 OmniDocBench v1.6 獲得 96.74 分,並在 OCRBench v2 英文及中文測試集分別取得 65.9 和 61.3 分。模型在多項文件解析、OCR、關鍵資訊提取、圖表説明及跨頁表格基準測試中取得領先成績。
Published on Oct 20
·
on Oct 28
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
OvisOCR3 is a compact 0.8B vision-language model that unifies diverse OCR tasks through multi-teacher distillation and reinforcement learning, achieving state-of-the-art results across document parsing, general OCR, key information extraction, figure captioning, and cross-page table reconstruction.
We introduce OvisOCR3, a 0.8B model for a broad range of OCR-related tasks. OvisOCR3 unifies document parsing, general OCR, key information extraction (KIE), document figure captioning, and cross-page table reconstruction in a single model. We train the model on real and synthetic data through continued pretraining and use domain-balanced mid-training to balance supervision across tasks with different data volumes and response lengths. Multi-teacher on-policy distillation transfers expertise from five domain-specific 4B teachers into the shared model, followed by mixed-task reinforcement learning with domain-specific rewards. We compare OvisOCR3 with general-purpose vision-language models and specialized OCR models on public and in-house benchmarks covering all five task families. In these comparisons, it achieves an overall score of 96.74 on OmniDocBench v1.6 and the highest overall scores on PureDocBench, Real5-OmniDocBench, and MDPBench. On OCRBench v2, it ranks first, scoring 65.9 and 61.3 on the English and Chinese splits, respectively. It also outperforms the baselines on OCRBench and all four KIE benchmarks. On DocFigCapBench, it achieves the highest localization F1 and a caption score close to that of Qwen3.8-Max. For cross-page tables, it leads in Relation F1 on CrossTabBench and truncated-table TEDS on MPDocBench-Parse. These results demonstrate that a compact 0.8B model can achieve strong performance across diverse OCR-related tasks. OvisOCR3 is available at https://huggingface.co/ATH-MaaS/OvisOCR3.
View arXiv page View PDF Add to collection
Models citing this paper 1
Image-Text-to-Text • 0.9B • Updated 1 day ago • 1.61k • 78ATH-MaaS/OvisOCR3
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.33066 in a dataset README.md to link it from this page.
Spaces citing this paper 2
Collections including this paper 1
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co