Fake News Detection System using XLNet model with Topic Distributions: CONSTRAINT@AAAI2021 Shared Task

With the ease of access to information, and its rapid dissemination over the\ninternet (both velocity and volume), it has become challenging to filter out\ntruthful information from fake ones. The research community is now faced with\nthe task of automatic detection of fake news, which carries real-world\nsocio-political impact. One such research contribution came in the form of the\nConstraint@AAA12021 Shared Task on COVID19 Fake News Detection in English. In\nthis paper, we shed light on a novel method we proposed as a part of this\nshared task. Our team introduced an approach to combine topical distributions\nfrom Latent Dirichlet Allocation (LDA) with contextualized representations from\nXLNet. We also compared our method with existing baselines to show that XLNet +\nTopic Distributions outperforms other approaches by attaining an F1-score of\n0.967.\n