RESEARCH PAPERSubmitted 11 Mar 2026

World2Act: Latent Action Post-Training from World Model Dynamics

An Dinh Vuong; Tuan Van Vo; Abdullah Sohail; Haoran Ding; Liang Ma; Xiaodan Liang; Anqing Duan; Ivan Laptev; Ian Reid

Read on arXiv PDF Project page

Classification

View four quadrants

Major category: WAMs
Quadrant: Outside quadrants
Architecture: Dual-system
Prediction paradigm: Other mechanisms
Subcategories: Policy post-training & WM-RL Generalization & action alignment
Source review status: Verified from primary sources

AT A GLANCE

Contribution

A contribution summary has not been added yet.

Abstract

An abstract has not been added yet.

Affiliations

Not listed in the collection.

BibTeX

@article{world2act,
      title={World2Act: Latent Action Post-Training from World Model Dynamics}, 
      author={An Dinh Vuong and Tuan Van Vo and Abdullah Sohail and Haoran Ding and Liang Ma and Xiaodan Liang and Anqing Duan and Ivan Laptev and Ian Reid},
      year={2026},
      journal={arXiv:2603.10422},
}

Classification

Contribution

Abstract

Affiliations

BibTeX

More in this category

DUET-DINO: Simultaneous Cross-View World Modeling for Latent Planning in Robot Manipulation

Grounding Generated Video Plans in Simulation Towards Versatile Dexterous Controllers

HaWMPO: Hallucination-Aware World Model-based Policy Optimization for Generalist Robot Policy