AI Paper Atlas

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

Lavida-O: Elastic Large Masked Diffusion Models for Unified Multimodal Understanding and Generation

Shufan Li, Jiuxiang Gu, Kangning Liu, Zhe Lin, Zijun Wei, Aditya Grover, Jason Kuen

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 34 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 planning (layout via bounding boxes) to generation (ablation)LaViDa-O · GenEvalaccuracy: 0.77 -> 0.85Improved
add self-reflection on top of planning (ablation)LaViDa-O · GenEvalaccuracy: 0.85 -> 0.89Improved
add planning and reflection (ablation)LaViDa-O · DPG-Benchscore: 81.8 -> 83.2Improved
add planning (identify edit region first) for editing (ablation)LaViDa-O · ImgEdit overallscore: 3.71 -> 3.80Improved
shrink generation branch (Elastic-MoT) vs equal-size MoT (ablation)Elastic-MoT · training per-step latencyspeedup: 2.23xImproved
decouple attention across modalities in last 16 layers (ablation)Elastic-MoT · training per-step latencyspeedup: 1.44x additionalImproved
decode grounding in 1 diffusion step instead of 4 (ablation)LaViDa-O · RefCOCO valPrecision@0.5: 92.3 -> 91.9Mixed
scale training data / retrain from LaViDa baseLaViDa-O vs LaViDa · ChartQAscore: 64.6 -> 80.0Improved
scale training data / retrain from LaViDa baseLaViDa-O vs LaViDa · MME-Pscore: 1366 -> 1431No clear effect

Later papers that compared with this method

2 later accepted papers report LaViDa-O as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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