AI Paper Atlas

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

Diagnosing and Correcting Concept Omission in Multimodal Diffusion Transformers

Kanghyun Baek, Jaihyun Lew, Chaehun Shin, Jungbeom Lee, Sungroh Yoon

ICML 2026 · Image generation › Text-to-image diffusion · approach: Multi-concept composition · 0 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 Omission Signal Intervention (OSI)FLUX.1-Dev · GenEval multi-object averageaccuracy: 0.47 -> 0.61Improved
add OSISD3.5-Medium · GenEval multi-object averageaccuracy: 0.55 -> 0.64Improved
add OSI to attribute tokens as well as object tokensFLUX.1-Dev · T2I-CompBench attribute binding (shape)score: 0.4995 -> 0.5819Improved
intervene on object tokens only instead of both object and attribute tokens (ablation)OSI on FLUX · T2I-CompBench attribute binding (color)score: 0.7923 -> 0.7678Hurt
reverse intervention direction (opposite) (ablation)OSI on FLUX · GenEval six-objectaccuracy: 0.18 -> 0.02Hurt
use random direction instead of learned omission direction (ablation)OSI on FLUX · GenEval six-objectaccuracy: 0.18 -> 0.30 (learned: 0.40)Mixed
select top-K probed heads instead of bottom-K heads (ablation)OSI on FLUX · GenEval five-objectaccuracy: 0.22 -> 0.40Improved
select top-K heads instead of random heads (ablation)OSI on FLUX · GenEval four-objectaccuracy: 0.55 -> 0.64Improved
select top-K heads instead of all 1368 heads (ablation)OSI on FLUX · GenEval four-objectaccuracy: 0.50 -> 0.64Mixed
apply OSI to the omitted token vs present tokens (ablation)OSI on FLUX · 100 failure casesomitted-object accuracy: 0.14 -> 0.70Improved
extend intervention from 0 to 5 early steps (ablation)OSI on FLUX · GenEval two-objectaccuracy: 0.82 -> 0.88Improved
extend intervention from 15 to 30 steps (ablation)OSI on FLUX · GenEval two-objectaccuracy: 0.92 -> 0.91No clear effect

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