AI Paper Atlas

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

OSPO: Object-Centric Self-Improving Preference Optimization for Text-to-Image Generation

Yoonjin Oh, Yongjin Kim, Hyomin Kim, Donghwan Chi, Sungwoong Kim

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Preference optimization for diffusion · 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 prompt densification before image generation (ablation)OSPO on Janus-Pro-7B · T2I-CompBench++ Attributescore: 0.641 -> 0.756Improved
add pair filtering and selection (with densification) (ablation)OSPO on Janus-Pro-7B · T2I-CompBench++ Attributescore: 0.716 -> 0.756Improved
add pair filtering alone (without densification) (ablation)OSPO on Janus-Pro-7B · T2I-CompBench++ Attributescore: 0.618 -> 0.549Hurt
add pair filtering and selection (without densification) (ablation)OSPO on Janus-Pro-7B · T2I-CompBench++ Attributescore: 0.618 -> 0.641Improved
add filtering and selection (with densification) (ablation)OSPO on Janus-Pro-7B · GenEval Overallscore: 0.813 -> 0.831Improved
add SFT loss to preference loss (ablation)OSPO on Janus-Pro-7B · GenEval Overallscore: 0.794 -> 0.831Improved
add SFT loss to preference loss (ablation)OSPO on Janus-Pro-7B · T2I-CompBench++ Attributescore: 0.776 -> 0.756Hurt
use object-weighted SimPO instead of standard SimPO (with SFT) (ablation)OSPO on Janus-Pro-7B · GenEval Positionscore: 0.795 -> 0.828Improved
use object-weighted SimPO instead of standard SimPO (with SFT) (ablation)OSPO on Janus-Pro-7B · T2I-CompBench++ Layoutscore: 0.437 -> 0.447Improved
apply OSPO to Janus-Pro-7BJanus-Pro-7B · T2I-CompBench++ Colorscore: 0.5215 -> 0.8567Improved
apply OSPO vs SUDER on CountingJanus-Pro-7B · GenEval Countingscore: 0.70 -> 0.63Hurt
increase training sample size and candidate pair count (ablation)OSPO · T2I-CompBench++ attribute and GenEvalscoreImproved

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