Hi @dunbar12138 🤗
I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work through Hugging Face's daily papers as yours got featured: https://huggingface.co/papers/2605.28763.
It's fantastic to see that you've already shared the CubePart model weights and a Gradio demo on the Hub! This makes it incredibly accessible for the community to explore your work.
I noticed in the paper that a major contribution is the construction of a large open-vocabulary, part-labeled 3D dataset with 462K assets. Would you like to host this dataset on https://huggingface.co/datasets as well?
Hosting on Hugging Face will give the dataset more visibility and enable better discoverability through metadata tags. It also allows researchers to easily load it via the datasets library:
from datasets import load_dataset
dataset = load_dataset("Roblox/CubePart-Dataset")Hugging Face also supports 3D model visualization in the dataset viewer, which would be a great way for people to preview the assets in the browser.
After uploading, we can also link the dataset to the paper page so that all your artifacts (models, demo, and data) are discoverable in one place.
Let me know if you're interested or need any guidance regarding the upload process!
Kind regards,
Niels
Hi @dunbar12138 🤗
I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work through Hugging Face's daily papers as yours got featured: https://huggingface.co/papers/2605.28763.
It's fantastic to see that you've already shared the CubePart model weights and a Gradio demo on the Hub! This makes it incredibly accessible for the community to explore your work.
I noticed in the paper that a major contribution is the construction of a large open-vocabulary, part-labeled 3D dataset with 462K assets. Would you like to host this dataset on https://huggingface.co/datasets as well?
Hosting on Hugging Face will give the dataset more visibility and enable better discoverability through metadata tags. It also allows researchers to easily load it via the
datasetslibrary:Hugging Face also supports 3D model visualization in the dataset viewer, which would be a great way for people to preview the assets in the browser.
After uploading, we can also link the dataset to the paper page so that all your artifacts (models, demo, and data) are discoverable in one place.
Let me know if you're interested or need any guidance regarding the upload process!
Kind regards,
Niels