AI Paper Atlas

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

FonTS: Text Rendering With Typography and Style Controls

Wenda Shi, Yiren Song, Dengming Zhang, Jiaming Liu, Xingxing Zou

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Layout-to-image control · 27 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
fine-tune only joint text attention (5% params) rather than text+image attention (ablation)TC-FT on Flux.1-dev · BTR-benchWord-Acc: 31.00 -> 55.00Improved
fine-tune only joint text attention rather than text+image attention (ablation)TC-FT on Flux.1-dev · BTR-benchOCR-Acc: 77.92 -> 82.85Improved
fine-tune T5 text encoder with new tokens (ablation)TC-FT on Flux.1-dev · BTR-benchOCR-Acc: 82.85 -> 0Hurt
train new tokens only without fine-tuning attention (ablation)TC-FT on Flux.1-dev · BTR-benchOCR-Acc: 82.85 -> 66.52Hurt
use enclosing token pair instead of single modifier token (ablation)ETC-tokens · BTR-benchWord-Acc: 32.00 -> 55.00Improved
use ETC-tokens instead of natural-language bold instruction (ablation)ETC-tokens · BTR-benchWord-Acc: 25.00 -> 55.00Improved
add second-phase Art-FT on SC-artext to style adapter (ablation)SCA · ATR-benchCLIP-I: 60.07 -> 65.12Improved
add TC-FT backbone fine-tuning (ablation)SCA without Art-FT · ATR-benchOCR-Acc: 28.89 -> 65.39Improved
add TC-FT backbone fine-tuning (ablation)SCA without Art-FT · ATR-benchCLIP-I: 60.07 -> 58.09Hurt
apply SCA to both MM-DiT and Single-DiT instead of MM-DiT only (ablation)SCA · ATR-benchAvg: 57.34 -> 62.17Improved
FonTS vs Glyph-ByT5 on OCR accuracy in basic text renderingBTR · BTR-benchOCR-Acc: 96.36 -> 82.85Hurt
use FonTS instead of Flux for scene textFlux.1-dev · MARIO-benchOCR-Acc: 24.17 -> 53.57Improved

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