Datasets accompanying the paper ConditionalQA: A Complex Reading Comprehension Dataset with Conditional Answers.
This dataset should ONLY be used for NLP research purpose. Answers are NOT verified by legal professionals and should NOT be used for any legal purposes.
Please generate your predictions using the format sample_output.json. Run the following command to evaluate your predictions with evaluate.py:
python evaluate.py --pred_file=sample_output.json --ref_file=v1_0/dev.json
Submit your predictions to the Leaderboard.
Please email your Codalab username to haitians@cs.cmu.edu if you would like your results to be added to the leaderboard. Include your organisation, a link to your paper, and a short description of your model in the email.
If you use these datasets please cite the following:
TBD