PulseAugur
EN
LIVE 10:03:34

Unlimited OCR 2 model achieves constant computational cost for parsing lengthy documents

By PulseAugur Editorial · [1 sources] ·

Researchers have introduced Unlimited OCR 2, a novel document parsing model designed to handle documents of virtually unlimited length without prohibitive memory overhead. This model upgrades decoder attention modules to Sparse Reference Sliding Window Attention (SR-SWA), which maintains constant computational costs by limiting active reference tokens to a fixed capacity. This approach enables significant speedups, achieving 7.0x and 5.6x per-step improvements over Full Attention and R-SWA respectively on hundred-page documents. Unlimited OCR 2 also demonstrates state-of-the-art performance on MPDocBench and MarathonDocBench, new benchmarks specifically designed for parsing lengthy documents. AI

IMPACT Enables efficient parsing of extremely long documents, potentially transforming how large-scale document analysis and transcription are performed.

RANK_REASON The cluster describes a new research paper detailing a novel model architecture and benchmark for document parsing. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Unlimited OCR 2 model achieves constant computational cost for parsing lengthy documents

How we ranked this

Signal score
1 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The cluster describes a new research paper detailing a novel model architecture and benchmark for document parsing. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
1 days old
Coverage has settled into its steady-state source set.

Full methodology in our editorial standards.

COVERAGE [1]

  1. arXiv cs.CL TIER_1 English(EN) · Shu Wei, Jingjing Wu, Qunyi Xie, Lingshu Zhang, Xiangcheng Du, Yannian Fu, Mengjie Wang, Zuyi Wang, Zhichao Liu, Yangliu Xu, Youyang Yin, Huanhuan Liu, YY, Yi Yang ·

    Unlimited OCR Works 2

    arXiv:2610.39308v2 Announce Type: replace Abstract: Unlimited OCR ushers the era of long-horizon parsing, substantially mitigating the growth of working memory as document lengthens. However, as document length scales to hundreds of pages, the visual context grows substantially, …

RELATED ENTITIES

RELATED TOPICS