Release MPAIL2 artifacts on Hugging Face

[NielsRogge]

Hi @UWRobotLearning 🤗

Niels here from the open-source team at Hugging Face. I discovered your work through Hugging Face's daily papers as your research on MPAIL2 was featured: https://huggingface.co/papers/2602.24121.

The paper page lets people discuss your work and find associated artifacts (like models or datasets). You can also claim the paper as yours, which will show it on your public profile at HF, and add your GitHub and project page URLs.

I noticed on your project page that the code for MPAIL2 is "coming soon." It would be fantastic to make the pre-trained world models and the demonstration datasets available on the 🤗 Hub once you are ready to release them. Hosting on Hugging Face improves discoverability, especially within the robotics community, and allows us to add metadata tags so people can find your work easily.

Uploading models

For robotics world models or policies, you can host checkpoints at https://huggingface.co/models. See here for a guide: https://huggingface.co/docs/hub/models-uploading. You can leverage the hf_hub_download one-liner to download checkpoints directly into your code.

Uploading datasets

It would also be great to make your task observations/demonstrations available on 🤗 , so that people can easily explore or use the data:

from datasets import load_dataset

dataset = load_dataset("your-hf-org-or-username/your-dataset")

See here for a guide: https://huggingface.co/docs/datasets/loading. Besides that, the dataset viewer allows people to quickly explore image-based trajectories directly in the browser.

Let me know if you're interested or need any help regarding this!

Cheers,

Niels
ML Engineer @ HF 🤗