mlsanityData to 14 Oct 2026

olmOCR 2: Unit Test Rewards for Document OCR

arXiv 2510.19817 ↗First posted 22 Oct 2025cs.CV cs.CLcited by 47

Not on the map. Why?

Abstract (arXiv)

We present olmOCR 2, the latest in our family of powerful OCR systems for converting digitized print documents, like PDFs, into clean, naturally ordered plain text. olmOCR 2 is powered by olmOCR-2-7B-1025, a specialized, 7B vision language model (VLM) trained using reinforcement learning with verifiable rewards (RLVR), where our rewards are a diverse set of binary unit tests. To scale unit test creation, we develop a pipeline for generating synthetic documents with diverse and challenging layouts, known ground-truth HTML source code, and extracted test cases. We show that RL training on these test cases results in state-of-the-art performance on olmOCR-Bench, our English-language OCR benchmark, with the largest improvements in math formula conversion, table parsing, and multi-column layouts compared to previous versions. We release our model, data and code under permissive open licenses.

Key facts. Cited by 47 papers in the map (data through 14 Oct 2026). Builds on 16 of its 26 references that are in the map. First posted on arXiv 22 Oct 2025.

Builds on

16 of 26 references.

References · 16

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2025Qwen2.5-VL Technical Reportcited by 5,587
  3. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  4. 2022Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timecited by 1,396
  5. 2024Tulu 3: Pushing Frontiers in Open Language Model Post-Trainingcited by 1,053
  6. 2021Merging Models with Fisher-Weighted Averagingcited by 621
  7. 2024MinerU: An Open-Source Solution for Precise Document Content Extractioncited by 267
  8. 2024General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Modelcited by 180
  9. 2025olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Modelscited by 125
  10. 2025MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsingcited by 116
  11. 2025PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Modelcited by 89
  12. 2024Image Over Text: Transforming Formula Recognition Evaluation with Character Detection Matchingcited by 41
4 more
  1. 2025Infinity Parser: Layout Aware Reinforcement Learning for Scanned Document Parsingcited by 17
  2. 2025Seeing is Believing? Mitigating OCR Hallucinations in Multimodal Large Language Modelscited by 12
  3. 2025DocR1: Evidence Page-Guided GRPO for Multi-Page Document Understandingcited by 11
  4. 2025DianJin-OCR-R1: Enhancing OCR Capabilities via a Reasoning-and-Tool Interleaved Vision-Language Modelcited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 43.

47 papers.

Sort:

By area:

  1. 2026LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCRcited by 23
  2. 2025Adaptation of Agentic AI: A Survey of Post-Training, Memory, and Skillscited by 22
  3. 2025MonkeyOCR v1.5 Technical Report: Unlocking Robust Document Parsing for Complex Patternscited by 13
  4. 2025Reading or Reasoning? Format Decoupled Reinforcement Learning for Document OCRcited by 9
  5. 2026HSD: Training-Free Acceleration for Document Parsing Vision-Language Models with Hierarchical Speculative Decodingcited by 6
  6. 2026Beyond NL2Code: A Structured Survey of Multimodal Code Intelligencecited by 6
  7. 2026MDPBench: A Benchmark for Multilingual Document Parsing in Real-World Scenarioscited by 5
  8. 2026OvisOCR2 Technical Reportcited by 5
  9. 2025Benchmarking Document Parsers on Mathematical Formula Extraction from PDFscited by 4
  10. 2026Orthogonal Model Mergingcited by 4
  11. 2026RTPrune: Reading-Twice Inspired Token Pruning for Efficient DeepSeek-OCR Inferencecited by 4
  12. 2026How Far Is Document Parsing from Solved? PureDocBench: A Source-Traceable Benchmark across Clean, Degraded, and Real-World Settingscited by 4
35 more
  1. 2026Infinity-Parser2 Technical Reportcited by 4
  2. 2026MonkeyOCRv2: A Visual-Text Foundation Model for Document AIcited by 4
  3. 2026GutenOCR: A Grounded Vision-Language Front-End for Documentscited by 2
  4. 2026Efficient Document Parsing via Parallel Token Predictioncited by 2
  5. 2026Beyond String Matching: Semantic Evaluation of PDF Table Extractioncited by 2
  6. 2026GlotOCR Bench: OCR Models Still Struggle Beyond a Handful of Unicode Scriptscited by 2
  7. 2026ZAYA1-VL-8B Technical Reportcited by 2
  8. 2026StrucTab: A Structured Optimization Framework for Table Parsingcited by 2
  9. 2026Benchmarking Vision-Language Models for French PDF-to-Markdown Conversioncited by 1
  10. 2026Synthetic Mixed Training: Scaling Parametric Knowledge Acquisition Beyond RAGcited by 1
  11. 2026TexOCR: Advancing Document OCR Models for Compilable Page-to-LaTeX Reconstructioncited by 1
  12. 2026METATR: A Multilingual, Evolving Benchmark for Automatic Text Recognitioncited by 1
  13. 2026Review Arcade: On the Human Alignment and Gameability of LLM Reviewscited by 1
  14. 2026Entropy-KL Divergence-based Token Masking: A Novel Approach for Selective Fine-tuning of Large Language Modelscited by 1
  15. 2026PerceptionRubrics: Calibrating Multimodal Evaluation to Human Perceptioncited by 1
  16. 2026Enhancing Large Multimodal Models in Key Information Extraction via Scene-Aware Document Synthesiscited by 1
  17. 2026daVinci-LLM:Towards the Science of Pretrainingcited by 0
  18. 2026JaWildText: A Benchmark for Vision-Language Models on Japanese Scene Text Understandingcited by 0
  19. 2026Parser-Oriented Structural Refinement for a Stable Layout Interface in Document Parsingcited by 0
  20. 2026DharmaOCR: Specialized Small Language Models for Structured OCR that outperform Open-Source and Commercial Baselinescited by 0
  21. 2026Zamba2-VL Technical Reportcited by 0
  22. 2026RealDocBench: A Benchmark for Field-Level QA and Layout Understanding on Real-World Regulated Documentscited by 0
  23. 2026Freeing the Law with LOCUS: A Local Ordinance Corpus for the United Statescited by 0
  24. 2026When Simpler Is Better: Evaluating Translation Pipelines for Medieval Latin Manuscriptscited by 0
  25. 2026OCR-Based Field Extraction for Archaeological Pottery Metadata: The CENTURIA Datasetcited by 0
  26. 2026Closing Cost-Quality Gap in Document VLMs: Difficulty-Aware Data Curation and Quality-Adjusted Deployment Economicscited by 0
  27. 2026WeVisDoc: From Coverage to Capability for Robust End-to-End Document Parsingcited by 0
  28. 2026PrismAlign: Prior-Steered Multi-View VLM Alignment for Hallucination-Robust Table OCRcited by 0
  29. 2026Layout-Guided Masking for GROBID: Lightweight Structural Gains in Large-Scale Scientific PDF Ingestioncited by 0
  30. 2026Orthogonal Yet Coupled: Decoupling Geometric Components for Model Mergingcited by 0
  31. 2026PolyOCR-Venus: Unified OCR Foundation Models for Text-Centric Visual Intelligencecited by 0
  32. 2026Improving OCR Faithfulness via Gated and Attenuated On-Policy Distillationcited by 0
  33. 2026HANS: A Handwritten Answer Sheet Dataset for Noisy Hybrid Document Parsingcited by 0
  34. 2026The Last Mile of Coding Agents for Long-Document Tasks: An Empirical Study on Clinical Trial Protocol Digitizationcited by 0
  35. 2026Equal CER, Unequal Damage: Severity-Weighted Character Error Rate for OCRcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024DocVLM: Make Your VLM an Efficient Reader
  2. 2026DODO: Discrete OCR Diffusion Models
  3. 2026Do VLMs Read or Rewrite? On Transcription Faithfulness in Vision-Language Models
  4. 2026Xiaomi-OCR-0 Technical Report
  5. 2024TAP-VL: Text Layout-Aware Pre-training for Enriched Vision-Language Models
  6. 2026LayoutLite: Token-Level Implicit Layout Analysis for Efficient Document OCR
  7. 2026From Retrieval to Recognition:How Vision--Language Models Become OCR Specialists
  8. 2026Beyond Blind Compliance: Benchmarking Task Verification in OCR Reasoning