DeltaV: Thinking with Visual State Updates in Unified Large Multimodal Models
Paper • 2607.08434 • Published • 2
Yuliang PRO
Leon5201314
AI & ML interests
Document-native intelligence.
Recent Activity
liked a model 15 days ago
wpj20000/DeltaV-2B upvoted a paper 15 days ago
DeltaV: Thinking with Visual State Updates in Unified Large Multimodal Models repliedto their post 17 days ago
0.7B MonkeyOCRv2 Outperforms Larger Models on 17-Language Document Parsing
MonkeyOCRv2-B-Parsing reaches 83.3 on MDPBench, a multilingual benchmark covering digital-born and photographed documents across 17 languages.
Results among evaluated open-source models:
• MonkeyOCRv2-B, 0.7B: 83.3
• dots.mocr, 3B: 80.5
• HunyuanOCR-1.5, 1B: 76.8
• PaddleOCR-VL-1.6, 0.9B: 75.0
• MinerU2.5-Pro, 1.2B: 71.0
The central idea is simple: before asking an LLM to reason over a document, the vision encoder must preserve every character stroke, digit, punctuation mark, and layout cue.
MonkeyOCRv2 is pretrained on 113M document images across 17 languages using joint image-to-text generation and pixel-level reconstruction.
Models:
https://huggingface.co/collections/zenosai/monkeyocrv2
Paper:
https://huggingface.co/papers/2607.11562
GitHub:
https://github.com/Yuliang-Liu/MonkeyOCRv2
Code and model weights are available under Apache-2.0.
We welcome tests on difficult multilingual, photographed, and visually ambiguous documents—especially failure cases.Organizations
upvoted a paper 15 days ago
upvoted a collection 20 days ago
upvoted a paper 20 days ago
upvoted a paper about 1 year ago