English

ABot-OCR Technical Report

Computer Vision and Pattern Recognition 2026-05-28 v1

Abstract

We introduce ABot-OCR, an end-to-end vision-language model that transcribes a page image directly into clean Markdown in a single forward pass. By doing so, our approach completely eliminates the need for brittle modular orchestration. To maximize parsing fidelity, we develop a dedicated data engine to provide large-scale, structurally consistent supervision. Furthermore, we propose Decoupled Heterogeneous Document Optimization, a structure-constrained reinforcement learning method that sharpens textual accuracy and strictly enforces markup well-formedness beyond supervised fine-tuning alone. Extensive evaluations demonstrate the superior performance of our framework. On the OmniDocBench v1.5 and v1.6 benchmarks, ABot-OCR achieves state-of-the-art scores of 92.81 and 93.30 among all end-to-end systems, substantially narrowing the performance gap relative to strong pipeline baselines. Finally, comprehensive multilingual text recognition across ten diverse languages further confirms the robust generalizability of ABot-OCR.

Keywords

Cite

@article{arxiv.2605.27978,
  title  = {ABot-OCR Technical Report},
  author = {Kaitao Jiang and Ruiyan Gong and Xiaolong Cheng and Kangning Niu and Tianlun Li and Mu Xu},
  journal= {arXiv preprint arXiv:2605.27978},
  year   = {2026}
}

Comments

21 pages, 11 figures, technical report