Taking a Stance on Fake News: Towards Automatic Disinformation Assessment via Deep Bidirectional Transformer Language Models for Stance Detection
This work addresses the global issue of disinformation on social media by providing an AI-based solution for stance detection, though it is incremental as it applies an existing transformer model to a specific task.
The paper tackles the problem of automatically detecting disinformation by using deep bidirectional transformer language models for stance detection, achieving a state-of-the-art weighted accuracy of 90.01% on the Fake News Challenge Stage 1 benchmark.
The exponential rise of social media and digital news in the past decade has had the unfortunate consequence of escalating what the United Nations has called a global topic of concern: the growing prevalence of disinformation. Given the complexity and time-consuming nature of combating disinformation through human assessment, one is motivated to explore harnessing AI solutions to automatically assess news articles for the presence of disinformation. A valuable first step towards automatic identification of disinformation is stance detection, where given a claim and a news article, the aim is to predict if the article agrees, disagrees, takes no position, or is unrelated to the claim. Existing approaches in literature have largely relied on hand-engineered features or shallow learned representations (e.g., word embeddings) to encode the claim-article pairs, which can limit the level of representational expressiveness needed to tackle the high complexity of disinformation identification. In this work, we explore the notion of harnessing large-scale deep bidirectional transformer language models for encoding claim-article pairs in an effort to construct state-of-the-art stance detection geared for identifying disinformation. Taking advantage of bidirectional cross-attention between claim-article pairs via pair encoding with self-attention, we construct a large-scale language model for stance detection by performing transfer learning on a RoBERTa deep bidirectional transformer language model, and were able to achieve state-of-the-art performance (weighted accuracy of 90.01%) on the Fake News Challenge Stage 1 (FNC-I) benchmark. These promising results serve as motivation for harnessing such large-scale language models as powerful building blocks for creating effective AI solutions to combat disinformation.