AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

UM-Text: A Unified Multimodal Model for Image Understanding and Visual Text Editing

Lichen Ma, Xiaolong Fu, , GaojingZhou, Zipeng Guo, Ting Zhu, Yichun Liu, Yu Shi, Jason Li, Junshi Huang

AAAI 2026 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 5 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
add character-level visual encoder (ablation)FLUX-Fill baseline · AnyText-benchmark Chinese (100k subset)Sen.ACC: 0.029 -> 0.676Improved
add character-level visual encoder (ablation)FLUX-Fill baseline · AnyText-benchmark English (100k subset)Sen.ACC: 0.309 -> 0.759Improved
add VLM embedding (ablation)UM-Text · AnyText-benchmark English (100k subset)Sen.ACC: 0.759 -> 0.782Improved
add VLM embedding (ablation)UM-Text · AnyText-benchmark Chinese (100k subset)Sen.ACC: 0.676 -> 0.698Improved
add latent-space regional consistency loss (ablation)UM-Text · AnyText-benchmark English (100k subset)Sen.ACC: 0.782 -> 0.799Improved
add RGB-space regional consistency loss (ablation)UM-Text · AnyText-benchmark Chinese (100k subset)Sen.ACC: 0.725 -> 0.746Improved
UM-Text vs FLUX-Text on LPIPSUM-Text · AnyText-benchmark ChineseLPIPS: 0.0487 -> 0.0481No clear effect
UM-Text vs DreamText on reconstruction LPIPSUM-Text · UDiffText benchmark ReconLPIPS: 0.0328 -> 0.0479Hurt
UM-Text vs DreamText on SeqAcc and FIDUM-Text · UDiffText benchmark ReconFID: 12.13 -> 6.57Improved
use UM-Designer layout/text with UM-Text vs AnyText-2UM-Text · UMT-benchmark ChineseSen.ACC: 0.720 -> 0.956Improved

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