Release medical negation benchmarks and NAST models on Hugging Face

[NielsRogge]

Hi @aliabbasi-ai 🤗

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/2602.12498.

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.

It'd be great to make the checkpoints and dataset benchmarks available on the 🤗 hub, to improve their discoverability/visibility within the medical AI community. We can add tags so that people find them when filtering https://huggingface.co/models and https://huggingface.co/datasets.

Uploading models

Would you like to host the NAST-adapted checkpoints (BioMedCLIP, etc.) on Hugging Face? Hosting on the hub enables better discoverability and allows users to use hf_hub_download or integrate with our libraries easily. See here for a guide: https://huggingface.co/docs/hub/models-uploading.

Uploading dataset

It would be awesome to make the MedNega-CXR benchmark and the contextual negation dataset available on 🤗 , so that people can do:

from datasets import load_dataset

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

Even if the raw MIMIC-CXR images require credentials, hosting the metadata, labels, and MCQs on Hugging Face allows for the dataset viewer to quickly explore the data in the browser. You can find a guide here: https://huggingface.co/docs/datasets/loading.

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

Cheers,

Niels
ML Engineer @ HF 🤗