This repo collects benchmark, papers, and codes about embodied world models.
If you find this repository useful, please consider giving us a star 🌟
-
DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge
arXiv25.07
[Paper] [Code] -
WorldVLA: Towards Autoregressive Action World Model
arXiv25.06
[Paper] [Code] -
Occupancy World Model for Robots
arXiv25.05
[Paper] -
Learning 3D Persistent Embodied World Models
arXiv25.05
[Paper] -
Cosmos-Transfer1: Conditional World Generation with Adaptive Multimodal Control
arXiv25.04
[Paper] -
TesserAct: Learning 4D Embodied World Models
arXiv25.04
[Paper] [Code] -
COMBO: Compositional World Models for Embodied Multi-Agent Cooperation
ICLR 2025
[Paper] [Code] -
Cosmos World Foundation Model Platform for Physical AI
arXiv25.03
[Paper] -
EnerVerse: Envisioning Embodied Future Space for Robotics Manipulation
arXiv25.02
[Paper] -
NavigateDiff: Visual Predictors are Zero-Shot Navigation Assistants
arXiv25.02
[Paper] -
GenEx: Generating an Explorable World
arXiv25.01
[Paper]
-
WHALE: Towards Generalizable and Scalable World Models for Embodied Decision-making
arXiv24.11
[Paper] -
MineDreamer: Learning to Follow Instructions via Chain-of-Imagination for Simulated-World Control
NIPS 2024 @ OWA
[Paper] [Code] -
UniSim: Learning Interactive Real-World Simulators
ICLR 2024
[Paper] -
IRASim: Learning Interactive Real-Robot Action Simulators
arXiv24.06
[Paper] [Code] -
RoboDreamer: Learning Compositional World Models for Robot Imagination
arXiv24.04
[Paper]