This is the dataset and code for our paper.
The dataset is located in the 'datasets' directory. We partitioned the dataset according to the standard 7:1:2 ratio. 'dataset_info.json' contains some statistical information about the dataset.
More detailed information can be found in the paper.
You can easily train using 'EduCDM/train.sh', just be mindful of the path issues.
cd EduCDM
sh train.sh
