AI Paper Atlas

Research map / Video / Video generation

Product of Experts for Visual Generation

Yunzhi Zhang, Carson Murtuza-Lanier, Zizhang Li, Yilun Du, Jiajun Wu

ICLR 2026 · Video › Video generation · approach: World-model video generation · 2 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 conditional sampling of generative experts (ablation)Ours · Image object insertion (graphics engine input, 30 scenes)MSE background: 1.511 -> 1.429Improved
add conditional sampling of generative experts (ablation)Ours · Image object insertion (graphics engine input, 30 scenes)GPT-4o controllability: 0.727 -> 0.827Improved
add conditional sampling of generative experts (ablation)Ours · Image object insertion (graphics engine input, 30 scenes)GPT-4o text alignment: 0.675 -> 0.817Improved
add conditional sampling of generative experts (ablation)Ours · Image object insertion (graphics engine input, 30 scenes)Aesthetic: 0.599 -> 0.596Hurt
increase particle count from L=1 to L=32Ours · MIG-Bench layout control (50 scenes)VQAScore: 0.567 -> 0.904Improved
increase particle count from L=1 to L=32Ours · MIG-Bench layout control (50 scenes)mIoU: 0.675 -> 0.728Improved
add SMC resampling at L=32 (ablation)Ours · MIG-Bench layout control (50 scenes)VQAScore: 0.854 -> 0.904Improved
add SMC resampling at L=32 (ablation)Ours · MIG-Bench layout control (50 scenes)mIoU: 0.691 -> 0.728Improved
add SMC resampling at L=4 (ablation)Ours · MIG-Bench layout control (50 scenes)mIoU: 0.715 -> 0.695Hurt
add SMC resampling at L=8 (ablation)Ours · MIG-Bench layout control (50 scenes)mIoU: 0.735 -> 0.715Hurt
add SMC resampling at L=8 (ablation)Ours · MIG-Bench layout control (50 scenes)VQAScore: 0.807 -> 0.879Improved
compose experts with Ours instead of Image2VWan2.1 14B · Physics-simulator video generation (object-centric simulation, 12 scenes)GPT-4o controllability: 0.708 -> 0.708No 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