Release reme.library (appworld.jsonl, bfcl_v3.jsonl) datasets on Hugging Face

[NielsRogge]

Hi @agentscope-ai 🤗

I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work on Arxiv and your exciting ReMe framework, featured on our papers page: https://huggingface.co/papers/2512.10696. The paper page lets people discuss your work and find related artifacts, and you can also claim the paper as yours to show on your Hugging Face profile, add GitHub and project page URLs.

I noticed that you've released the reme.library dataset, including appworld.jsonl and bfcl_v3.jsonl, alongside your code on GitHub. It would be fantastic to make these datasets available on the 🤗 Hub as well, to significantly improve their discoverability and visibility.

Hosting them on Hugging Face allows others to easily access and use your data with:

from datasets import load_dataset

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

You can find a guide for uploading datasets here: https://huggingface.co/docs/datasets/loading. We also provide a dataset viewer that allows users to quickly explore the data in their browser.

After uploading, we can link your datasets to the paper page (read here) so people can easily find them.

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

Cheers,

Niels
ML Engineer @ HF 🤗

[jinliyl]

Hi Niels,

Thanks for reaching out! We're glad to see the ReMe framework featured on the Hugging Face papers page.

That’s a great suggestion—hosting the datasets on the Hub will definitely make them more accessible. We’ll look into the upload process and get back to you once they're live to finalize the link with the paper page.

Thanks again for the support.

Best,

jinli.yl / @agentscope-ai team