RoBBR is a risk-of-bias benchmark with three tasks: Main Task: Risk-of-Bias Determination, Support Sentence Retrieval (SSR), and Support Judgment Selection (SJS).
You can read more detailed description of the dataset in RoBBR Github
We recommand you to can download the datasets using the following commands:
git clone
https://huggingface.co/datasets/RoBBR-Benchmark/RoBBR
cp -r RoBBR/*.json dataset/
rm -r RoBBR
