mlsanityData to 14 Oct 2026

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

arXiv 2609.37712 ↗First posted 29 Sep 2026cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Optical Character Recognition (OCR) is evolving from plain-text transcription toward general visual intelligence, requiring models to recognize, localize, and reason over textual information in complex visual environments. However, existing OCR systems often excel at only some tasks and struggle to balance recognition, parsing, and reasoning across scenarios. In this report, we present PolyOCR, a family of unified OCR foundation models of varying scales. PolyOCR combines a shared instruction-following framework with a large-scale data engine that converts heterogeneous visual resources into quality-verified OCR supervision. We introduce Competence-Guided Policy Optimization, which combines verifier-based Group Relative Policy Optimization with on-policy distillation through sample-wise routing based on teacher reliability and the teacher--student competence gap. We also introduce OCRBench v2.1, our revision of OCRBench v2 with manually verified annotation corrections and task-aligned scoring metrics. Extensive experiments across OCRBench v2.1, CC-OCR, in-house KIE Benchmark, OmniDocBench v1.6 and MDPBench demonstrate that PolyOCR achieves state-of-the-art or highly competitive performance.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 88 of its 106 references that are in the map. First posted on arXiv 29 Sep 2026.

Builds on

88 of 106 references.

References · 88

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2017On Calibration of Modern Neural Networkscited by 4,960
  4. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  5. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  6. 2025Qwen3-VL Technical Reportcited by 2,793
  7. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  8. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,919
  9. 2019Towards VQA Models That Can Readcited by 1,718
  10. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 1,614
  11. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,510
  12. 2022ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoningcited by 1,479
38 more
  1. 2020DocVQA: A Dataset for VQA on Document Imagescited by 1,317
  2. 2020Gradient Surgery for Multi-Task Learningcited by 1,256
  3. 2026DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligencecited by 1,040
  4. 2023On-Policy Distillation of Language Models: Learning from Self-Generated Mistakescited by 939
  5. 2017GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networkscited by 834
  6. 2024MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?cited by 586
  7. 2019LayoutLM: Pre-training of Text and Layout for Document Image Understandingcited by 553
  8. 2015An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognitioncited by 520
  9. 2022LayoutLMv3: Pre-training for Document AI with Unified Text and Image Maskingcited by 485
  10. 2023OCRBench: On the Hidden Mystery of OCR in Large Multimodal Modelscited by 476
  11. 2023Large Language Models Are State-of-the-Art Evaluators of Translation Qualitycited by 420
  12. 2021OCR-free Document Understanding Transformercited by 407
  13. 2023DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretrainingcited by 399
  14. 2016Synthetic Data for Text Localisation in Natural Imagescited by 396
  15. 2019Which Tasks Should Be Learned Together in Multi-task Learning?cited by 375
  16. 2019PubLayNet: largest dataset ever for document layout analysiscited by 351
  17. 2019FUNSD: A Dataset for Form Understanding in Noisy Scanned Documentscited by 349
  18. 2019Scene Text Visual Question Answeringcited by 331
  19. 2021TrOCR: Transformer-based Optical Character Recognition with Pre-trained Modelscited by 279
  20. 2021ICDAR2019 Competition on Scanned Receipt OCR and Information Extractioncited by 256
  21. 2017EAST: An Efficient and Accurate Scene Text Detectorcited by 248
  22. 2023Nougat: Neural Optical Understanding for Academic Documentscited by 245
  23. 2019Image-based table recognition: data, model, and evaluationcited by 237
  24. 2016COCO-Text: Dataset and Benchmark for Text Detection and Recognition in Natural Imagescited by 236
  25. 2024General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Modelcited by 180
  26. 2024TextMonkey: An OCR-Free Large Multimodal Model for Understanding Documentcited by 174
  27. 2024OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoningcited by 173
  28. 2017Total-Text: A Comprehensive Dataset for Scene Text Detection and Recognitioncited by 162
  29. 2024OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotationscited by 150
  30. 2022Scene Text Recognition with Permuted Autoregressive Sequence Modelscited by 149
  31. 2025PaddleOCR 3.0 Technical Reportcited by 143
  32. 2021PubTables-1M: Towards comprehensive table extraction from unstructured documentscited by 126
  33. 2023Skill-it! A Data-Driven Skills Framework for Understanding and Training Language Modelscited by 121
  34. 2025MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsingcited by 116
  35. 2018Scene Text Detection and Recognition: The Deep Learning Eracited by 114
  36. 2025Ovis2.5 Technical Reportcited by 108
  37. 2022Towards End-to-End Unified Scene Text Detection and Layout Analysiscited by 97
  38. 2025PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Modelcited by 89

50 most cited of 88 All 88 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025DianJin-OCR-R1: Enhancing OCR Capabilities via a Reasoning-and-Tool Interleaved Vision-Language Model
  2. 2026From Retrieval to Recognition:How Vision--Language Models Become OCR Specialists
  3. 2026Where Vision Becomes Text: Locating the OCR Routing Bottleneck in Vision-Language Models
  4. 2026PP-OCRv5: A Specialized 5M-Parameter Model Rivaling Billion-Parameter Vision-Language Models on OCR Tasks
  5. 2025Consensus Entropy: Harnessing Multi-VLM Agreement for Self-Verifying and Self-Improving OCR
  6. 2026SP-DocReader: Difference-Aware Self-Play for Precise Document OCR
  7. 2025Benchmarking Vision-Language Models on Optical Character Recognition in Dynamic Video Environments
  8. 2026DocOCR-Eval: A Correction-Based Framework for OCR Tool Selection Without Ground Truth