AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

UniDoc: Unified Pretraining Framework for Document Understanding

Jiuxiang Gu, Jason Kuen, Vlad I Morariu, Handong Zhao, Rajiv Jain, Nikolaos Barmpalios, Ani Nenkova, Tong Sun

NeurIPS 2021 · Vision-language and multimodal › Human and embodied modeling · approach: Layout-aware document understanding · 45 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Combine vision and language instead of using either stream alone (ablation)UDoc without pretraining · FUNSDF1: 77.49, 78.46 -> 83.34Improved
Add VCL to MSM, training the visual backbone end-to-end (ablation)UDoc pretraining · FUNSDF1: 84.37 -> 86.87Improved
Add VCL to MSM, training the visual backbone end-to-end (ablation)UDoc pretraining · CORD†F1: 97.44 -> 98.70Improved
Add VCL to MSM, training the visual backbone end-to-end (ablation)UDoc pretraining · RVL-CDIPAccuracy: 93.10 -> 93.59Improved
Add VLA to MSM + VCL (ablation)UDoc pretraining · FUNSDF1: 86.87 -> 87.38Improved
Add VLA to MSM + VCL (ablation)UDoc pretraining · CORD†F1: 98.70 -> 98.75Improved
Add VLA to MSM + VCL (ablation)UDoc pretraining · RVL-CDIPAccuracy: 93.59 -> 93.92Improved
Add image-text matching (REL) to MSM + VCL + VLA (ablation)UDoc pretraining · FUNSDF1: 87.38 -> 87.20Hurt
Add image-text matching (REL) to MSM + VCL + VLA (ablation)UDoc pretraining · CORD†F1: 98.75 -> 98.13Hurt
Add image-text matching (REL) to MSM + VCL + VLA (ablation)UDoc pretraining · RVL-CDIPAccuracy: 93.92 -> 93.64Hurt
Add masked language modeling to UDoc pretraining (ablation)UDoc · FUNSDF1: 87.38 -> 83.76Hurt
Add masked language modeling to UDoc pretraining (ablation)UDoc · CORDF1: 98.75 -> 98.63Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy