AI Paper Atlas

Research map / 3D vision / 3D generation

MVISTA-4D: View-Consistent 4D World Model with Test-Time Action Inference for Robotic Manipulation

Jiaxu Wang, JIANG Yicheng, Tianlun HE, Jingkai SUN, Qiang Zhang, Jiahang Cao, Zesen Gan, Mingyuan Sun, Qiming Shao, Xiangyu Yue

ICML 2026 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 0 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 geometry-aware deformable cross-view attention (vs none) (ablation)MVISTA-4D · RoboTwin 4D generationAbRel (x10^2): 3.34 -> 2.60Improved
use deformable cross-view attention instead of simple epipolar attention (ablation)MVISTA-4D · RoboTwin 4D generationEMD (x10^2): 12.50 -> 9.90Improved
add cross-modality fusion between RGB and depth (ablation)MVISTA-4D · RoboTwin 4D generationPSNR: 20.16 -> 22.91Improved
optimize trajectory latent at test time instead of using action-head prior (ablation)MVISTA-4D · RLBench manipulationsuccess rate (%): 72.5 -> 72.6Improved
add residual IDM on top of decoded trajectory prior (ablation)MVISTA-4D · RLBench manipulationsuccess rate (%): 69.0 -> 72.6Improved
use residual IDM instead of full point-based IDM (ablation)MVISTA-4D · RLBench manipulationsuccess rate (%): 68.8 -> 72.6Improved
add cross-view and cross-modality fusion (ablation)MVISTA-4D · RLBench manipulationsuccess rate (%): 66.5 -> 72.6Improved
add TCN-VAE trajectory latent (ablation)MVISTA-4D · RoboTwin manipulationsuccess rate (%): 41.5 -> 43.0Improved
fuse depth via token-level modality fusion instead of channel concatenation (ablation)MVISTA-4D · RLBench manipulationsuccess rate (%): 66.1 -> 72.6Improved
initialize trajectory latent with action-head prediction (+ActionInit) (ablation)MVISTA-4D · RLBench manipulationsuccess rate (%): 72.6 -> 76.5Improved
+ActionInit reduces optimization to 10-15 steps (ablation)MVISTA-4D · RLBench inference latencytotal time (s): 58s -> 47sImproved
use MVISTA-4D instead of UniPi* for appearanceRLBench 4D generation · RLBenchPSNR: 23.88 -> 23.31Hurt

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