AdaWorld: Learning Adaptable World Models with Latent Actions
Paper
•
2503.18938
•
Published
•
2
TL;DR: We pretrain AdaWorld with continuous latent actions from thousands of environments, making it highly efficient for action transfer, world model adaptation, and visual planning, even with minimal interactions and finetuning steps.
Check out https://github.com/Little-Podi/AdaWorld
@article{gao2025adaworld,
title={AdaWorld: Learning Adaptable World Models with Latent Actions},
author={Gao, Shenyuan and Zhou, Siyuan and Du, Yilun and Zhang, Jun and Gan, Chuang},
journal={arXiv preprint arXiv:2503.18938},
year={2025}
}
If you have any questions or comments, feel free to contact me through email sygao@connect.ust.hk