AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

UniMERNet: A Universal Network for Real-World Mathematical Expression Recognition

Zhuangcheng Gu, Guang Liang, Bin Wang, Zhiyuan Zhao, Qintong Zhang, Weijia Li, Chao Xu, Bo Zhang, Botian Shi, Jiang Wu, Wentao Zhang, Conghui He

CVPR 2026 · Architectures and efficiency › Attention and transformers · approach: State space and recurrent sequence models · 63 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
replace global attention with Raster-Scan attention (ablation)UniMERNet encoder · UniMER-Test CPECDM: 0.943 -> 0.952Improved
replace global attention with Raster-Scan attention (ablation)UniMERNet encoder · encoder efficiencyFPS (bs=1): 2.11 -> 10.52Improved
replace window attention with Raster-Scan attention (ablation)UniMERNet encoder · UniMER-Test CPECDM: 0.931 -> 0.952Improved
use horizontal-then-vertical instead of vertical-then-horizontal order (ablation)R-S Attention · UniMER-Test CPECDM: 0.948 -> 0.952Improved
add ConvEnhance module (ablation)UniMERNet encoder · UniMER-Test CPECDM: 0.952 -> 0.955Improved
add ConvEnhance module (ablation)UniMERNet encoder · UniMER-Test SCECDM: 0.937 -> 0.939No clear effect
train on UniMER-1M instead of Pix2tex+HWE (ablation)UniMERNet · UniMER-Test CPECDM: 0.665 -> 0.955Improved
train on UniMER-1M instead of Pix2tex+HWE (ablation)UniMERNet · UniMER-Test SPECDM: 0.983 -> 0.991Improved
add handwritten data to Pix2tex training (ablation)UniMERNet · UniMER-Test CPECDM: 0.667 -> 0.665No clear effect
increase input resolution to 384x1344UniMERNet · UniMER-Test CPECDM: 0.955 -> 0.972Improved
compare UniMERNet with PosFormer on HME100KUniMERNet · HME100KExpRate: 69.5 -> 68.5No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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