Release GeoDANO, GeoCLIP models and geometric benchmark on Hugging Face

[NielsRogge]

Hi @ml-postech 🤗

Niels here from 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/2502.11360.
The paper page lets people discuss about your paper and lets them find artifacts about it (your models, datasets or demo for instance), you can also claim
the paper as yours which will show up on your public profile at HF, add Github and project page URLs.

I saw in your abstract that "The implementation is available at this https URL" for your GeoDANO and GeoCLIP models, as well as the new benchmark dataset for evaluating geometric feature recognition. We're very excited about your work!

It'd be great to make these checkpoints and the dataset available on the 🤗 hub once they are publicly released, to improve their discoverability/visibility.
We can add tags so that people find them when filtering https://huggingface.co/models and https://huggingface.co/datasets.

Uploading models

See here for a guide: https://huggingface.co/docs/hub/models-uploading.

In this case, we could leverage the PyTorchModelHubMixin class which adds from_pretrained and push_to_hub to any custom nn.Module. Alternatively, one can leverages the hf_hub_download one-liner to download a checkpoint from the hub.
For GeoDANO and GeoCLIP, which are geometric vision-language models for solving plane geometry problems, the image-text-to-text pipeline tag would be highly relevant.

We encourage researchers to push each model checkpoint to a separate model repository, so that things like download stats also work. We can then also link the checkpoints to the paper page.

Uploading dataset

Would be awesome to make your new geometric feature recognition benchmark dataset available on 🤗 , so that people can do:

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.
For a dataset containing geometric diagram-caption pairs, the image-text-to-text task category would be appropriate.

Besides that, there's the dataset viewer which allows people to quickly explore the first few rows of the data in the browser.

Let me know if you're interested/need any help regarding this, once the artifacts are publicly released!

Cheers,

Niels
ML Engineer @ HF 🤗