Release MedSAM-Agent trajectory datasets on Hugging Face

[NielsRogge]

Hi @Saint-lsy 🤗

I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work on Arxiv regarding MedSAM-Agent 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 your work and find associated artifacts. It's great to see that you've already hosted the MedSAM-Agent-Qwen3-VL-8B-MedSAM2 model checkpoint on the Hub! You can claim the paper as yours on the HF paper page, which will show up on your public profile, and add your GitHub and project page URLs.

I noticed in your README that you are planning to release the SFT trajectory and RL training datasets soon. Would you like to host these datasets on Hugging Face as well?

Hosting on Hugging Face will give your data more visibility and enable better discoverability through our task tags. It also allows the community to easily explore and load the data using the datasets library:

from datasets import load_dataset

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

We also provide a dataset viewer that allows people to explore the trajectories directly in their browser. If you're interested, you can find a guide on uploading datasets here.

After they are uploaded, we can link both the model and the datasets to the paper page so people can discover the full workflow of your research.

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

Kind regards,

Niels

[Saint-lsy]

Thanks for your interest in our work! We will open-source the dataset in the future.