README in progress

Usage

The dataset is in folder according to the task and type (e.g. generation or collection for IR).

You can load a specific file (say the test set of the generation task) with this command:

from datasets import load_dataset
dataset = load_dataset("jhu-clsp/CLERC", data_files={"data": f"generation/test.jsonl"})["data"]

Change the values in data_files to match the file you want to load.