mlsanityData to 7 Oct 2026

Xiaomi-OCR-0 Technical Report

arXiv 2609.36136 ↗First posted 28 Sep 2026cs.CVcited by 0

On the map: Multimodal LLMs / vision-language models → Document understanding and multimodal RAG

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Document understanding and multimodal RAG, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 61 of its 75 references that are in the map. First posted on arXiv 28 Sep 2026.

Builds on

61 of 75 references.

Closest ancestors · 17

Same direction, newest first

  1. 2026HunyuanOCR-1.5: Making Lightweight OCR VLMs Faster and Better6 Jul
  2. 2026ABot-OCR Technical Report27 May
  3. 2026Q-Mask: Query-driven Causal Masks for Text Anchoring in OCR-Oriented Vision-Language Models31 Mar
  4. 2026Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training25 Mar
  5. 2026Qianfan-OCR: A Unified End-to-End Model for Document Intelligence11 Mar
  6. 2026Real5-OmniDocBench: A Full-Scale Physical Reconstruction Benchmark for Robust Document Parsing in the Wild4 Mar
  7. 2026OCRVerse: Towards Holistic OCR in End-to-End Vision-Language Models29 Jan
  8. 2026LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR20 Jan
  9. 2025dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Model2 Dec
  10. 2025Infinity Parser: Layout Aware Reinforcement Learning for Scanned Document Parsing17 Oct
  11. 2025PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model16 Oct
  12. 2025MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing26 Sep
5 more
  1. 2025Logics-Parsing Technical Report24 Sep
  2. 2025POINTS-Reader: Distillation-Free Adaptation of Vision-Language Models for Document Conversion1 Sep
  3. 2025A Token-level Text Image Foundation Model for Document Understanding4 Mar
  4. 2025olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Models25 Feb
  5. 2023OCRBench: On the Hidden Mystery of OCR in Large Multimodal Models13 May

Other references · 44

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  2. 2025Qwen2.5-VL Technical Reportcited by 5,418
  3. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  4. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,678
  5. 2025Qwen3-VL Technical Reportcited by 2,534
  6. 2019Towards VQA Models That Can Readcited by 1,696
  7. 2022ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoningcited by 1,449
  8. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,419
  9. 2020DocVQA: A Dataset for VQA on Document Imagescited by 1,288
  10. 2025OpenAI GPT-5 System Cardcited by 910
  11. 2019LayoutLM: Pre-training of Text and Layout for Document Image Understandingcited by 549
  12. 2026Kimi K2.5: Visual Agentic Intelligencecited by 449
32 more
  1. 2025GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learningcited by 439
  2. 2021OCR-free Document Understanding Transformercited by 400
  3. 2019Scene Text Visual Question Answeringcited by 327
  4. 2019Image-based table recognition: data, model, and evaluationcited by 233
  5. 2026Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Rankingcited by 211
  6. 2025MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipecited by 174
  7. 2025DeepSeek-OCR: Contexts Optical Compressioncited by 174
  8. 2025PaddleOCR 3.0 Technical Reportcited by 136
  9. 2025Ovis2.5 Technical Reportcited by 103
  10. 2025MonkeyOCR: Document Parsing with a Structure-Recognition-Relation Triplet Paradigmcited by 81
  11. 2025Dolphin: Document Image Parsing via Heterogeneous Anchor Promptingcited by 78
  12. 2026DeepSeek-OCR 2: Visual Causal Flowcited by 68
  13. 2026MOPD: Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Trainingcited by 60
  14. 2026PaddleOCR-VL-1.5: Towards a Multi-Task 0.9B VLM for Robust In-the-Wild Document Parsingcited by 54
  15. 2026Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexitycited by 48
  16. 2026Seed1.8 Model Card: Towards Generalized Real-World Agencycited by 47
  17. 2024UniMERNet: A Universal Network for Real-World Mathematical Expression Recognitioncited by 41
  18. 2020A Survey of Deep Learning Approaches for OCR and Document Understandingcited by 32
  19. 2024TextHawk2: A Large Vision-Language Model Excels in Bilingual OCR and Grounding with 16x Fewer Tokenscited by 32
  20. 2026MinerU2.5-Pro: Pushing the Limits of Data-Centric Document Parsing at Scalecited by 26
  21. 2026GLM-OCR Technical Reportcited by 24
  22. 2026Dolphin-v2: Universal Document Parsing via Scalable Anchor Promptingcited by 20
  23. 2026PaddleOCR-VL-1.6: Expanding the Frontier of Document Parsing with Under-Optimized Region Refinement and Progressive Post-Trainingcited by 20
  24. 2025UniRec-0.1B: Unified Text and Formula Recognition with 0.1B Parameterscited by 12
  25. 2026FireRed-OCR Technical Reportcited by 12
  26. 2026Youtu-Parsing: Perception, Structuring and Recognition via High-Parallelism Decodingcited by 9
  27. 2025Multimodal Large Language Models for Text-rich Image Understanding: A Comprehensive Reviewcited by 8
  28. 2026Unlimited OCR Workscited by 8
  29. 2026Chronicles-OCR: A Cross-Temporal Perception Benchmark for the Evolutionary Trajectory of Chinese Characterscited by 5
  30. 2026OvisOCR2 Technical Reportcited by 5
  31. 2026MonkeyOCRv2: A Visual-Text Foundation Model for Document AIcited by 4
  32. 2026RT-DocLayout: Real-Time End-to-End Document Layout Analysis with Reading Order in the Wildcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond Blind Compliance: Benchmarking Task Verification in OCR Reasoning
  2. 2026Multilingual OCR-Aware Fine-Tuning and Prompt-Guided Chain-of-Thought Reasoning for Multimodal Large Language Models
  3. 2026OCR-MetaReasoning Benchmark: Evaluating the Meta-Reasoning Ability of MLLMs in Text-Rich Image Understanding
  4. 2025DianJin-OCR-R1: Enhancing OCR Capabilities via a Reasoning-and-Tool Interleaved Vision-Language Model
  5. 2026PP-OCRv5: A Specialized 5M-Parameter Model Rivaling Billion-Parameter Vision-Language Models on OCR Tasks
  6. 2024OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning
  7. 2026Where Vision Becomes Text: Locating the OCR Routing Bottleneck in Vision-Language Models
  8. 2024TAP-VL: Text Layout-Aware Pre-training for Enriched Vision-Language Models