AI Paper Atlas

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

ViMo: A Generative Visual GUI World Model for App Agents

Dezhao Luo, Bohan Tang, Kang Li, Georgios Papoudakis, Jifei Song, Shaogang Gong, Jianye HAO, Jun Wang, Kun Shao

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Layout-to-image control · 39 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 ViMo visual world model to T3AT3A · step accuracystep accuracy: 43.13 -> 49.20Improved
add ViMo visual world model to M3AM3A · step accuracystep accuracy: 46.01 -> 50.16Improved
add ViMo vs language-based world model Change-textM3A · step accuracystep accuracy: 47.28 -> 50.16Improved
add language-based HTML-text world modelM3A · step accuracystep accuracy: 46.01 -> 46.65No clear effect
add ViMo in online navigationT3A (Gemini-2.0-Flash) · AndroidWorld task accuracytask accuracy: 33.19 -> 40.95Improved
add ViMo to unseen Apps zero-shotM3A (Gemini-2.0-Flash) · step accuracystep accuracy: 44.2 -> 47.6Improved
decouple graphic (STR) from text generationViMo vs IP2P* · GUI quality automatics_h: 0.69 -> 0.76Improved
ViMo vs HTML-vision on instructional accuracyHTML-vision · GUI quality automatics_ia: 85.77 -> 75.39Hurt
ViMo vs IP2P* in user studyIP2P* · GUI quality user studys_h: 0.63 -> 0.88Improved
retain static text as pixels; use action instruction conditioning (ablation)ViMo · step accuracystep accuracyImproved
increase rollout iterations beyond one (ablation)ViMo · step accuracystep accuracyMixed
longer trajectories (T+1 to T+4)ViMo · trajectory synthesistrajectory accuracy: 57.89 -> 12.28Hurt

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