The ClaimBuster dataset consists of statements extracted from all U.S. general election presidential debates (1960-2016) along with human-annotated check-worthiness labels where each sentence is categorized into one of the three categories: non-factual statement, unimportant factual statement, and check-worthy factual statement. If you use this dataset, please cite the following papers: @inproceedings{arslan2020claimbuster,     title={{A Benchmark Dataset of Check-worthy Factual Claims}},     author={Arslan, Fatma and Hassan, Naeemul and Li, Chengkai and Tremayne, Mark },     booktitle={14th International AAAI Conference on Web and Social Media},     year={2020},     organization={AAAI} } @article{meng2020gradient,   title={Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims},   author={Meng, Kevin and Jimenez, Damian and Arslan, Fatma and Devasier, Jacob Daniel and Obembe, Daniel and Li, Chengkai},   journal={arXiv preprint arXiv:2002.07725},   year={2020} }