百度开源 Unlimited OCR 2，以 SR-SWA 机制支持超长文档解析
百度发布并开源 Unlimited OCR 2，通过 SR-SWA 机制将文本与视觉注意力交互压缩为固定大小的 working references，在有限上下文窗口内解析近乎任意长度的文档。随着文档增长，其单步解码延迟和内存占用接近恒定，并在 MPDocBench 和 MarathonDocBench 上达到 SOTA 性能，覆盖从数十页到数百页的文档。
Unlimited OCR 2 is here, and it's open source. 📖 ✨
Longer documents can mean a lot more attention computation. Our new SR-SWA mechanism keeps that in check by condensing text-visual attention interactions into a fixed-size set of working references.
Here's what it delivers:
- Document parsing at virtually any length, within a finite context window
- Nearly constant per-step decoding latency and memory footprint as documents grow longer
- State-of-the-art performance on MPDocBench and MarathonDocBench, covering documents from dozens to hundreds of pages
Check out the model here ↓
来源：Baidu_Inc · x.com