AI Paper Atlas

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

DreamOmni2: Multimodal Instruction-based Generation and Editing

Bin Xia, bohao peng, Yuechen Zhang, Junjia Huang, Jiyang Liu, Jingyao Li, Haoru Tan, Sitong Wu, Chengyao Wang, Yitong Wang, Bei Yu, Jiaya Jia

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Text generation and structured prediction

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
train generation/editing LoRA on the synthetic multimodal data (no VLM) (ablation)Flux Kontext · Editing, concrete objectssuccess rate: 0.1220 -> 0.3659Improved
train generation/editing LoRA on the synthetic multimodal data (no VLM) (ablation)Flux Kontext · Editing, abstract attributessuccess rate: 0.0122 -> 0.3171Improved
add VLM training only (no editing model training) (ablation)Flux Kontext · Generation, concrete objectssuccess rate: 0.3750 -> 0.5417Improved
add joint training of VLM and editing model over training the editing model alone (ablation)DreamOmni2 editing · Editing, concrete objectssuccess rate: 0.3659 -> 0.6585Improved
add joint training of VLM and editing model over training the editing model alone (ablation)DreamOmni2 editing · Editing, abstract attributessuccess rate: 0.3171 -> 0.6280Improved
add joint training of VLM and generation model over training the generation model alone (ablation)DreamOmni2 generation · Generation, abstract attributessuccess rate: 0.3444 -> 0.6333Improved
add joint training over VLM-only training (ablation)DreamOmni2 editing · Editing, concrete objectssuccess rate: 0.2439 -> 0.6585Improved
add index encoding (with position shift) (ablation)DreamOmni2 · Editing, concrete objectssuccess rate: 0.4634 -> 0.6585Improved
add index encoding (without position shift) (ablation)DreamOmni2 · Editing, concrete objectssuccess rate: 0.2439 -> 0.3415Improved
add position encoding shift (with index encoding) (ablation)DreamOmni2 · Editing, concrete objectssuccess rate: 0.3415 -> 0.6585Improved
add position encoding shift (with index encoding) (ablation)DreamOmni2 · Generation, abstract attributessuccess rate: 0.4556 -> 0.6333Improved
use DreamOmni2 vs Nano Banana for abstract attribute editingDreamOmni2 · Editing, abstract attributes (human)success rate: 0.3293 -> 0.6829Improved

Later papers that compared with this method

4 later accepted papers report DreamOmni2 as a baseline in their results tables; 4 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