AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Transferability Between Understanding and Generation in Unified Multimodal Models

Jiwon Kang, Heeji Yoon, Jaewoo Jung, Jaewon Min, Minkyeong Jeon, Biyeon Hwang, Sangwon Jung, Seungryong Kim

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 4 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
train on counting understanding only (und->gen transfer) (ablation)Lumina-DiMOO · Counting generation (GenEval-style)accuracy (%): 48.0 -> 57.0Improved
train on counting understanding only (und->gen) (ablation)Janus-Pro · Counting generation (GenEval-style)accuracy (%): 38.0 -> 32.0Hurt
train on counting understanding only (und->gen) (ablation)BAGEL · Counting generation (GenEval-style)accuracy (%): 42.0 -> 47.0Improved
train on counting understanding only (und->gen) (ablation)BLIP3-o · Counting generation (GenEval-style)accuracy (%): 31.0 -> 37.0Improved
train on counting generation only (gen->und) (ablation)Lumina-DiMOO · Counting understanding (PixMo-Count)accuracy (%): 22.0 -> 30.0Improved
train on counting generation only (gen->und) (ablation)BLIP3-o · Counting understanding (PixMo-Count)accuracy (%): 63.0 -> 62.0No clear effect
train counting via understanding instead of direct generation fine-tuningLumina-DiMOO · Counting generation (GenEval-style)FID vs baseline images: 52.51 -> 31.47Improved
train counting via understanding instead of direct generation fine-tuningLumina-DiMOO · Counting generation (GenEval-style)Inception Score: 15.29 -> 17.55Improved
train counting via understanding instead of direct generation fine-tuningLumina-DiMOO · Counting generation (GenEval-style)MAD: 0.91 -> 0.83Improved
train spatial relation via understanding instead of generationLumina-DiMOO · Spatial relation generation (diagonal, synthetic)accuracy (%): 80.0 -> 74.0Hurt
train spatial relation via understanding instead of generationLumina-DiMOO · Spatial relation generation (diagonal, synthetic)FID vs baseline images: 32.28 -> 30.95Improved
train text via understanding instead of generationLumina-DiMOO · Text generation (rendered Markdown)BLEU: 0.354 -> 0.274Hurt

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