Release AlphaPFN on Hugging Face

Hi @herilalaina 🤗

I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work on Arxiv and was wondering whether you would like to submit it to hf.co/papers to improve its discoverability. If you are one of the authors, you can submit it at https://huggingface.co/papers/submit.

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

Would you like to host the $\alpha$-PFN model checkpoints you've pre-trained on https://huggingface.co/models? I noticed in your GitHub repository that you have released checkpoints for PES, MES, and JES that download automatically. Hosting these on Hugging Face will give your work more visibility and enable better discoverability. We can add tags in the model cards so that people find the models easier, link them directly to the paper page, and track usage stats.

If you're down, I'm leaving a guide here. For your architecture, you can use the PyTorchModelHubMixin class which adds from_pretrained and push_to_hub functionality to your model class, allowing people to download and use your models right away. Alternatively, you can simply use hf_hub_download to fetch the checkpoints programmatically within your existing from_pretrained logic.

Let me know if you're interested or need any guidance!

Kind regards,

Niels

Review the existing checkpoint download and from_pretrained logic, then read Hugging Face's model-uploading guide and the linked PyTorchModelHubMixin documentation. Clarify which AlphaPFN checkpoints and model-card links should be hosted; the work is done when the agreed checkpoints are available on Hugging Face and can be discovered and downloaded as intended.