Regularly updated OpenReview dataset annotated with metadata.
from datasets import load_dataset
To see the available metadata see the data model here
No full papers available due to licensing -> but can be easily obtained as follows:
for sample in dataset["train"]:
if sample["openreview_submission_id"] is not None:… See the full description on the dataset page:
https://huggingface.co/datasets/nhop/OpenReview.
