Baidu Inc.· @Baidu_Inc · X·· 2026-06-23AI 评分71
AI 导读
百度开源Unlimited OCR，专为一次性读取长文档设计。模型总参数量3B，仅激活500M，在OmniDocBench v1.5和v1.6上取得端到端SOTA。核心创新为参考滑动窗口注意力（R-SWA），模拟人类抄书过程，保持源、近期上下文和后续焦点，同时软遗忘无关信息。凭借恒定KV缓存大小和更低注意力成本，可在单次前向传播中转录40+页，不丢失上下文也不减速。模型已开源至GitHub和Hugging Face。
正文 · 原文
3B total parameters & 500M activated, yet powerful enough to transcribe 40+ pages in one pass while keeping context intact. Meet Unlimited OCR!
We’re open-sourcing Unlimited OCR — built to read long documents in one pass. With 3B total parameters and only 500M activated, Unlimited OCR sets new end-to-end SOTA results on OmniDocBench v1.5 and v1.6. The key innovation is Reference Sliding Window Attention (R-SWA), inspired by how humans transcribe books: keeping the source, recent context, and next words in focus, while softly forgetting what’s no longer needed. With constant KV Cache size and lower attention cost, Unlimited OCR can transcribe 40+ pages in a single forward pass — without losing context or slowing down. Explore the model👇: --GitHub: https://github.com/baidu/Unlimited-OCR --Hugging Face: https://huggingface.co/baidu/Unlimited-OCR在 X 查看被引用的帖子
来源：Baidu Inc. · x.com