AI Paper Atlas

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

Discovering Divergent Representations between Text-to-Image Models

Lisa Dunlap, Joseph E. Gonzalez, Trevor Darrell, Fabian Caba Heilbron, Josef Sivic, Bryan Russell

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 2 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
run five evolutionary iterations instead of one for description discovery (ablation)CompCon · ID2 (Top 1)description score: 0.59 -> 0.64Improved
run five iterations instead of one (Top 5) (ablation)CompCon · ID2 (Top 5)description score: 0.72 -> 0.78Improved
CompCon attribute discovery instead of VisDiffattribute discovery · ID2 (Top 1)attribute score: 0.47 -> 0.60Improved
CompCon attribute discovery instead of TF-IDFattribute discovery · ID2 (Top 1)attribute score: 0.23 -> 0.60Improved
CompCon instead of LLM-only end-to-endCompCon · ID2 (Top 1)description score: 0.03 -> 0.64Improved
use IDEFICS open VLM instead of GPT-4o VLM (ablation)CompCon · ID2 (Top 1)attribute score: 0.60 -> 0.56Hurt
use IDEFICS plus llama3-8b instead of GPT-4o for both (ablation)CompCon · ID2 (Top 1)attribute score: 0.60 -> 0.39Hurt
use IDEFICS plus llama3-8b instead of GPT-4o for both (ablation)CompCon · ID2 (Top 1)description score: 0.64 -> 0.46Hurt
inject 25% VLM errors in attribute discovery (ablation)CompCon · ID2 (Top 1)attribute score: 0.60 -> 0.56No clear effect
more evolutionary iterations raise proportion of diverging prompts (Mandala) (ablation)CompCon · Mandala diverging promptsproportion diverging: 12% -> 76%Improved
more iterations (Wet Streets) (ablation)CompCon · Wet Streets diverging promptsproportion diverging: 8% -> 52%Improved
more iterations (Decay) (ablation)CompCon · Decay diverging promptsproportion diverging: 8% -> 44%Improved

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