RLA-WM predicts residual latent actions via flow matching to create visual feature world models that outperform prior feature-based and diffusion approaches while enabling offline video-based robot RL.
Title resolution pending
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
fields
cs.CV 6years
2026 6verdicts
UNVERDICTED 6representative citing papers
MultiWorld is a scalable framework for multi-agent multi-view video world models that improves controllability and consistency over single-agent baselines in game and robot tasks.
ACWM-Phys benchmark shows action-conditioned world models generalize on simple geometric interactions but drop sharply on deformable contacts, high-dimensional control, and complex articulated motion, indicating reliance on visual appearance over learned physics.
DiGSeg repurposes diffusion U-Nets as generalist segmentation learners by conditioning on image-mask latents and multi-scale CLIP text features, achieving strong cross-domain performance.
PhyEdit improves physical accuracy in image object manipulation by using explicit geometric simulation as 3D-aware guidance combined with joint 2D-3D supervision.
Semantic latent spaces from pretrained encoders outperform reconstruction-based spaces for robotic world models on planning and downstream policy performance.
citing papers explorer
-
Learning Visual Feature-Based World Models via Residual Latent Action
RLA-WM predicts residual latent actions via flow matching to create visual feature world models that outperform prior feature-based and diffusion approaches while enabling offline video-based robot RL.
-
MultiWorld: Scalable Multi-Agent Multi-View Video World Models
MultiWorld is a scalable framework for multi-agent multi-view video world models that improves controllability and consistency over single-agent baselines in game and robot tasks.
-
ACWM-Phys: Investigating Generalized Physical Interaction in Action-Conditioned Video World Models
ACWM-Phys benchmark shows action-conditioned world models generalize on simple geometric interactions but drop sharply on deformable contacts, high-dimensional control, and complex articulated motion, indicating reliance on visual appearance over learned physics.
-
Diffusion Model as a Generalist Segmentation Learner
DiGSeg repurposes diffusion U-Nets as generalist segmentation learners by conditioning on image-mask latents and multi-scale CLIP text features, achieving strong cross-domain performance.
-
PhyEdit: Towards Real-World Object Manipulation via Physically-Grounded Image Editing
PhyEdit improves physical accuracy in image object manipulation by using explicit geometric simulation as 3D-aware guidance combined with joint 2D-3D supervision.
-
Reconstruction or Semantics? What Makes a Latent Space Useful for Robotic World Models
Semantic latent spaces from pretrained encoders outperform reconstruction-based spaces for robotic world models on planning and downstream policy performance.