Make dataset available on HF

[NielsRogge]

Hi @AoiDragon,

Niels here from the open-source team at Hugging Face. Congrats on your work and the ECCV Oral! Discovered it from here: https://huggingface.co/papers/2403.09792 (feel free to claim it as an author).

Would you be interested in providing the dataset on the hub, rather than Google Drive, to improve its usability/discoverability? We can then also link it to the paper page. So that people can do:

from datasets import load_dataset

dataset = load_dataset("your-hf-username-or-organization/hades-images")

See here for a guide: https://huggingface.co/docs/datasets/image_dataset (you can then call dataset.push_to_hub("your-hf-username-or-organization/hades-images") to push it to the hub).

This would make the dataset easier accessible, and also discoverable since you can add metadata tags to a dataset card, like "jailbreaking".

Let us know if you need any help.

Cheers,

Niels
ML Engineer @ HF 🤗

[AoiDragon]

Hi @NielsRogge,

Thank you and the Hugging Face team for recognizing our work! We've recently organized our data as an evaluation benchmark for MLLMs, and we're excited to upload our dataset to Hugging Face soon.

[NielsRogge]

Great, let me know when it would be available :)

[AoiDragon]

Our benchmark is available on https://huggingface.co/datasets/Monosail/HADES.

[NielsRogge]

Awesome :) also opened https://huggingface.co/datasets/Monosail/HADES/discussions/1 to link it to the paper.

Would it be possible to mention the 🤗 Dataset in the README of this Github repo?

[AoiDragon]

Sure, thanks for the reminder.