2 datasets found
  1. P

    WHOOPS! Dataset

    • paperswithcode.com
    • library.toponeai.link
    Updated Oct 29, 2023
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    Nitzan Bitton-Guetta; Yonatan Bitton; Jack Hessel; Ludwig Schmidt; Yuval Elovici; Gabriel Stanovsky; Roy Schwartz (2023). WHOOPS! Dataset [Dataset]. https://paperswithcode.com/dataset/whoops
    Explore at:
    Dataset updated
    Oct 29, 2023
    Authors
    Nitzan Bitton-Guetta; Yonatan Bitton; Jack Hessel; Ludwig Schmidt; Yuval Elovici; Gabriel Stanovsky; Roy Schwartz
    Description

    WHOOPS! Is a dataset and benchmark for visual commonsense. The dataset is comprised of purposefully commonsense-defying images created by designers using publicly-available image generation tools like Midjourney. It contains commonsense-defying image from a wide range of reasons, deviations from expected social norms and everyday knowledge.

  2. h

    WEIRD

    • huggingface.co
    Updated Jun 3, 2025
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    MERA (2025). WEIRD [Dataset]. https://huggingface.co/datasets/MERA-evaluation/WEIRD
    Explore at:
    Dataset updated
    Jun 3, 2025
    Authors
    MERA
    License

    Attribution 4.0 (CC BY 4.0)https://creativecommons.org/licenses/by/4.0/
    License information was derived automatically

    Description

    WEIRD

    This dataset is used in the paper Through the Looking Glass: Common Sense Consistency Evaluation of Weird Images.

      Task description

    WEIRD is an extended version of a binary classification subtask of the original English WHOOPS! benchmark. The dataset evaluates the ability to detect violations of commonsense. Commonsense violations are situations that contradict the norm of reality. For example, penguins can't fly, children don't drive cars, guests don't set the table… See the full description on the dataset page: https://huggingface.co/datasets/MERA-evaluation/WEIRD.

  3. Not seeing a result you expected?
    Learn how you can add new datasets to our index.

Share
FacebookFacebook
TwitterTwitter
Email
Click to copy link
Link copied
Close
Cite
Nitzan Bitton-Guetta; Yonatan Bitton; Jack Hessel; Ludwig Schmidt; Yuval Elovici; Gabriel Stanovsky; Roy Schwartz (2023). WHOOPS! Dataset [Dataset]. https://paperswithcode.com/dataset/whoops

WHOOPS! Dataset

Explore at:
Dataset updated
Oct 29, 2023
Authors
Nitzan Bitton-Guetta; Yonatan Bitton; Jack Hessel; Ludwig Schmidt; Yuval Elovici; Gabriel Stanovsky; Roy Schwartz
Description

WHOOPS! Is a dataset and benchmark for visual commonsense. The dataset is comprised of purposefully commonsense-defying images created by designers using publicly-available image generation tools like Midjourney. It contains commonsense-defying image from a wide range of reasons, deviations from expected social norms and everyday knowledge.

Search
Clear search
Close search
Google apps
Main menu