# DeepSeek-OCR 2: Visual Causal Flow

arXiv: https://arxiv.org/abs/2601.20552 · HTML: https://mlsanity.com/paper/2601.20552/

## Key facts

- Cited by 74 papers in the map (data through 14 Oct 2026).
- Builds on 41 of its 48 references that are in the map.
- First posted on arXiv 28 Jan 2026.

## Builds on (most cited first, 30 of 41)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [YOLO9000: Better, Faster, Stronger](https://mlsanity.com/paper/1612.08242/) (2016)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [Multilingual Denoising Pre-training for Neural Machine Translation](https://mlsanity.com/paper/2001.08210/) (2020)
- [Chameleon: Mixed-Modal Early-Fusion Foundation Models](https://mlsanity.com/paper/2405.09818/) (2024)
- [DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models](https://mlsanity.com/paper/2512.02556/) (2025)
- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [MinerU: An Open-Source Solution for Precise Document Content Extraction](https://mlsanity.com/paper/2409.18839/) (2024)
- [OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalization](https://mlsanity.com/paper/2212.12017/) (2022)
- [Patch n' Pack: NaViT, a Vision Transformer for any Aspect Ratio and Resolution](https://mlsanity.com/paper/2307.06304/) (2023)
- [DeepSeek-OCR: Contexts Optical Compression](https://mlsanity.com/paper/2510.18234/) (2025)
- [General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model](https://mlsanity.com/paper/2409.01704/) (2024)
- [OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotations](https://mlsanity.com/paper/2412.07626/) (2024)
- [olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Models](https://mlsanity.com/paper/2502.18443/) (2025)
- [PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model](https://mlsanity.com/paper/2510.14528/) (2025)

## Built on by (most cited first, 30 of 74)

- [Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs](https://mlsanity.com/paper/2605.00674/) (2026)
- [Multimodal OCR: Parse Anything from Documents](https://mlsanity.com/paper/2603.13032/) (2026)
- [HunyuanOCR-1.5: Making Lightweight OCR VLMs Faster and Better](https://mlsanity.com/paper/2607.04884/) (2026)
- [FireRed-OCR Technical Report](https://mlsanity.com/paper/2603.01840/) (2026)
- [OCRVerse: Towards Holistic OCR in End-to-End Vision-Language Models](https://mlsanity.com/paper/2601.21639/) (2026)
- [UniVLR: Unifying Text and Vision in Visual Latent Reasoning for Multimodal LLMs](https://mlsanity.com/paper/2605.11856/) (2026)
- [Unlimited OCR Works](https://mlsanity.com/paper/2606.23050/) (2026)
- [Penguin-VL: Exploring the Efficiency Limits of VLM with LLM-based Vision Encoders](https://mlsanity.com/paper/2603.06569/) (2026)
- [AgenticOCR: Parsing Only What You Need for Efficient Retrieval-Augmented Generation](https://mlsanity.com/paper/2602.24134/) (2026)
- [U-VLM: Hierarchical Vision Language Modeling for Report Generation](https://mlsanity.com/paper/2603.00479/) (2026)
- [Logics-Parsing-Omni Technical Report](https://mlsanity.com/paper/2603.09677/) (2026)
- [MDPBench: A Benchmark for Multilingual Document Parsing in Real-World Scenarios](https://mlsanity.com/paper/2603.28130/) (2026)
- [Reading or Guessing? Visual Grounding Failures of Vision-Language Models for OCR in Ancient Greek Editions](https://mlsanity.com/paper/2605.27750/) (2026)
- [OvisOCR2 Technical Report](https://mlsanity.com/paper/2607.13639/) (2026)
- [Data Science and Technology Towards AGI Part I: Tiered Data Management](https://mlsanity.com/paper/2602.09003/) (2026)
- [SnapGuard: Lightweight Prompt Injection Detection for Screenshot-Based Web Agents](https://mlsanity.com/paper/2604.25562/) (2026)
- [RTPrune: Reading-Twice Inspired Token Pruning for Efficient DeepSeek-OCR Inference](https://mlsanity.com/paper/2605.00392/) (2026)
- [How Far Is Document Parsing from Solved? PureDocBench: A Source-Traceable Benchmark across Clean, Degraded, and Real-World Settings](https://mlsanity.com/paper/2605.07492/) (2026)
- [Infinity-Parser2 Technical Report](https://mlsanity.com/paper/2607.07836/) (2026)
- [MonkeyOCRv2: A Visual-Text Foundation Model for Document AI](https://mlsanity.com/paper/2607.11562/) (2026)
- [Source or It Didn't Happen: A Multi-Agent Framework for Citation Hallucination Detection](https://mlsanity.com/paper/2605.08583/) (2026)
- [MolEmb: Multimodal Large Language Models Can Be Strong Molecular Embedding Models](https://mlsanity.com/paper/2608.23646/) (2026)
- [LingDT-VL-OCR: Structure-Aware Document-Level Parsing with Fine-Grained Visual Reference](https://mlsanity.com/paper/2603.11044/) (2026)
- [P-MTP: Efficient Document Parsing via Multi-Token Prediction with Progressive Depth Scaling](https://mlsanity.com/paper/2606.24447/) (2026)
- [Advancing WordArt-Oriented Scene Text Recognition: Datasets and Methods](https://mlsanity.com/paper/2606.24484/) (2026)
- [Why Struggle with Continuous Latents? Interpretable Discrete Latent Reasoning via Rendered Compression](https://mlsanity.com/paper/2606.29712/) (2026)
- [StrucTab: A Structured Optimization Framework for Table Parsing](https://mlsanity.com/paper/2606.29905/) (2026)
- [Do VLMs Read or Rewrite? On Transcription Faithfulness in Vision-Language Models](https://mlsanity.com/paper/2607.21617/) (2026)
- [Towards Universal Khmer Text Recognition](https://mlsanity.com/paper/2603.00702/) (2026)
- [Towards Khmer Scene Document Layout Detection](https://mlsanity.com/paper/2603.00707/) (2026)

## Abstract (arXiv)

We present DeepSeek-OCR 2 to investigate the feasibility of a novel encoder-DeepEncoder V2-capable of dynamically reordering visual tokens upon image semantics. Conventional vision-language models (VLMs) invariably process visual tokens in a rigid raster-scan order (top-left to bottom-right) with fixed positional encoding when fed into LLMs. However, this contradicts human visual perception, which follows flexible yet semantically coherent scanning patterns driven by inherent logical structures. Particularly for images with complex layouts, human vision exhibits causally-informed sequential processing. Inspired by this cognitive mechanism, DeepEncoder V2 is designed to endow the encoder with causal reasoning capabilities, enabling it to intelligently reorder visual tokens prior to LLM-based content interpretation. This work explores a novel paradigm: whether 2D image understanding can be effectively achieved through two-cascaded 1D causal reasoning structures, thereby offering a new architectural approach with the potential to achieve genuine 2D reasoning. Codes and model weights are publicly accessible at http://github.com/deepseek-ai/DeepSeek-OCR-2.