Skip to content
Preprint

Xiaomi-OCR-0 Technical Report

Sep 2026 · 0 citations · 75 references
Computer Science

TL;DR

An approximately 170M-sample OCR-centric corpus is built using an automated data engine that combines expert consensus, render-based verification, and targeted synthesis and shows that, with sufficient parsing training, OCR-centric understanding supervision provides additional gains for document parsing.

Abstract

Compact OCR-specific vision-language models achieve strong document parsing performance, but often rely on costly supervision and focus primarily on visual-text reconstruction. We introduce Xiaomi-OCR-0, a unified 0.8B model for document parsing and OCR-centric understanding. We build an approximately 170M-sample OCR-centric corpus using an automated data engine that combines expert consensus, render-based verification, and targeted synthesis. Starting from Qwen3.5-0.8B, our progressive training recipe combines Q-Mask-based text anchoring, continued pretraining, and mixed-task reinforcement learning (Mix-RL). Xiaomi-OCR-0 achieves 95.24 on Real5-OmniDocBench, 96.83 on OmniDocBench v1.6, and 87.94 on Wild-OmniDocBench, while reaching an average score of 83.2 across five OCR-oriented VQA benchmarks. Ablations further show that, with sufficient parsing training, OCR-centric understanding supervision provides additional gains for document parsing. Homepage: https://huggingface.co/spaces/SeerRay-Lab/Xiaomi-OCR-0.

View source

Similar papers

#artificial intelligence Preprint Sep 2026

PolyOCR-Venus: Unified OCR Foundation Models for Text-Centric Visual Intelligence

This report presents PolyOCR, a family of unified OCR foundation models of varying scales that combines a shared instruction-following framework with a large-scale data engine that converts heterogeneous visual resources into quality-verified OCR supervision.

Guang-Zhan Huang, Yong-Shuo Zhang, Bing-Tao Fu et al. · 0 citations
Preprint Oct 2026

Xiaomi-Lens-0 Technical Report

Multimodal large language models (MLLMs) answer questions about whole images well, yet they still struggle with fine-grained region understanding. A key reason is that text-only supervision never specifies which regions support an answer. To close this gap, we present Xiaomi-Lens-0, a 4B MLLM built on Qwen3.5 that pred...

Pei Fu, Yi-Ke Liu, Yu-Xin Sun et al. · 0 citations
#computer vision Preprint Sep 2026

Jina-OCR-v1: Efficient Document Parsing with Speculative Decoding and Dense Verifiable Rewards

Jina-OCR-v1 is presented, an end-to-end document parsing model built to serve on low-budget GPUs that combines the compressed-vision encoder and the 3B mixture-of-experts decoder of DeepSeek-OCR with a FastMTP speculative decoding head that shares a single draft block recursively across K=3 prediction steps.

Alejandro Barón García, Feng Wang, Emília Garcia Casademont et al. · 0 citations
Preprint Sep 2026

From Retrieval to Recognition:How Vision--Language Models Become OCR Specialists

Comparisons show that OCR specialization largely preserves head identity, retaining 17-20 of the top 20 heads per task with all-head rank correlations of 0.874-0.942, while redistributing their functional and causal strengths.

Yuan-Xiang Huangfu, Han-Meng Zhong, Linqing Chen et al. · 0 citations
#artificial intelligence Preprint Oct 2026

SP-DocReader: Difference-Aware Self-Play for Precise Document OCR

Accurate page transcription remains difficult for vision language models under limited input and training budgets. We present SP-DocReader, a self-play framework for optical character recognition (OCR) that targets residual errors after supervised fine-tuning. Reading Discrepancy Masking aligns reference and generated...

Wen-Jie Liao, Xiao-Hui Song, Liang Zhao et al. · 0 citations
Conference Oct 2026

SyncDoc: a synergistic evolution framework for structured document understanding

SyncDoc is developed, a synergistic evolution framework built upon LayoutLMv3, which introduces three complementary modules to enhance fine-grained visual structures, a Dynamic Masking Strategy to reduce dependence on textual semantics, and a Deep Cross-modal Orientation module to improve consistency between textual an...

Jia-Xuan Song, Gang He · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.