AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Divide and Conquer: Decoupled Representation Alignment for Multimodal World Models

Junyuan Xiao, Dingkang Liang, Xin Zhou, Yixuan Ye, Tongtong Su, Guangmo Yi, Bin Xia, Qiang Lyu, Shurui Shi, Jun Huang, Jianlou Si, Wenming Yang

ECCV 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal fusion with missing modalities · 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 M2-REPA (decoupled multi-expert alignment) to RGB-DM baselineRGB-DM Baseline · RealEstate10K 8-frameFVD: 102.44 -> 81.13Improved
add M2-REPA to RGB-DM baselineRGB-DM Baseline · RealEstate10K 200-frameFVD: 407.50 -> 262.50Improved
add M2-REPA to RGB-DM baselineRGB-DM Baseline · Minecraft 150-frameFVD: 191.25 -> 81.81Improved
naively align DINOv2, SAM2 and DepthAnythingV2 simultaneously without decoupling vs best single REPA (ablation)REPA · RealEstate10K 200-frameFVD: 301.25 -> 328.75Hurt
add modality-specific decoupling (CKA loss) to multi-expert alignment (ablation)REPA (D+S+DA) · RealEstate10K 200-frameFVD: 328.75 -> 262.50Improved
use CKA decoupling loss instead of cos^2 loss (ablation)M2-REPA · RealEstate10K 200-frameFVD: 319.75 -> 262.50Improved
use CKA decoupling loss instead of cos^2 loss (ablation)M2-REPA · RealEstate10K 8-frameFVD: 99.56 -> 81.13Improved
replace attention-based adapter projector with lightweight MLP (ablation)M2-REPA · RealEstate10K 200-frame (projection head ablation)FVD: 312.50 -> 262.50Improved
use 3-layer MLP head instead of 2-layer (ablation)M2-REPA · RealEstate10K 200-frame (projection head depth)FVD: 355.25 -> 262.50Improved
increase MLP head depth from 3 to 9 layers (ablation)M2-REPA · RealEstate10K 200-frame (projection head depth)FVD: 262.50 -> 258.75No clear effect
lower decoupling weight from 200 to 0.05 (ablation)M2-REPA · RealEstate10K 200-frame (decoupling weight sweep)FVD: 328.75 -> 262.50Improved
add M2-REPA to TesserAct (CogVideoX-5B)TesserAct · RealEstate10K (TesserAct CogVideoX-5B)FVD: 639.3 -> 418.3Improved

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