AI Paper Atlas

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

PosterVerse: A Full-Workflow Framework for Commercial-Grade Poster Generation with HTML-Based Scalable Typography

Junle Liu, Peirong Zhang, Yuyi Zhang, Pengyu Yan, Hui Zhou, Xinyue Zhou, Fengjun Guo, Lianwen Jin

AAAI 2026 · Image generation › Text-to-image diffusion · approach: Layout-to-image control · 3 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 hierarchical dynamic prompt sampling (basic, medium, detailed) in background training (ablation)Flux.1-dev LoRA · Background generationFID: 136.72 -> 62.54Improved
add hierarchical dynamic prompt sampling in background training (ablation)Flux.1-dev LoRA · Background generationIS: 62.39 -> 77.85Improved
add DIPR fine-tuned blueprint model, basic requirement input (ablation)PosterVerse vs Seedream3.0 · User studyaverage score: 4.92 -> 6.76Improved
add DIPR fine-tuned blueprint model, basic requirement input (ablation)PosterVerse vs GPT-4o · User studyaverage score: 4.52 -> 6.76Improved
use detailed instead of basic requirement inputSeedream3.0 · User studyaverage score: 4.92 -> 6.12Improved
use detailed instead of basic requirement inputPosterVerse · User studyaverage score: 6.76 -> 6.85No clear effect
use PosterVerse instead of GPT-4ofull poster pipeline · PosterDNA test setCorrect Rate (CR): 49.73% -> 92.33%Improved
use PosterVerse instead of GPT-4ofull poster pipeline · PosterDNA test setF1: 48.49% -> 78.58%Improved
use PosterVerse HTML layout rendering instead of GPT-4o static image outputfull poster pipeline · PosterDNA test setOverlap: 0.0106 -> 0.0027Improved
use PosterVerse instead of Seedream3.0full poster pipeline · PosterDNA test setGPT-4o evaluation average: 7.82 -> 8.02Improved
use fine-tuned Qwen2.5-14B blueprint model instead of Claude 3.7 Sonnetblueprint creation · Blueprint stageEdit Distance: 22.32 -> 12.00Improved
use LoRA-tuned Flux.1 with hierarchical prompts instead of Flux.1 devbackground generation · Background stageFID: 40.65 -> 26.8Improved

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