Welcome to the CodaLab page for the JEC-QA. JEC-QA is a question-answering dataset collected from the National Judicial Examination of China. It contains 26,365 multiple-choice and multiple-answer questions in total. The task of the dataset is to predict the answer using the questions and relevant articles. You can get the dataset and more details from the main page of JEC-QA.
The submissions are evaluated using accuracy on test dataset of single/all KD/CA questions. There will be six scores which denote the accuracy on single-answer KD-questions (kds), all KD-questions (kda), single-answer CA-questions (cas), all KD-questions (kda), single-answer questions (as), all questions (aa). The evaluation script is available on Github.
You should strictly follow the rules that we mention on mainpage of JEC-QA.
Start: Dec. 18, 2019, midnight
You must be logged in to participate in competitions.Sign In