Beyond the Horizon: Using Mixture of Experts for Domain Agnostic Fake News Detection

In recent years, social media have become one of the main means to quickly spread information worldwide, but this rapid dissemination also brings significant risks of misinformation and fake news, which can cause widespread confusion, erode public trust, and contribute to social and political instability. This scenario is further exacerbated by the fact that fake news can span various topics across different domains, making it impracticable for a single moderator to manage the massive quantity of data. The use of Machine Learning, particularly language models, is rising as an effective solution to mitigate the risk of misinformation. However, a single model cannot fully capture the complexity and variety of the information it needs to process, often failing to classify examples from new domains. In this work, the aforementioned challenges are addressed by leveraging a novel hierarchical deep-ensemble framework. This framework aims to integrate various domains to offer enhanced predictions for new ones. Specifically, the approach involves learning a distinct model for each domain and refining them through domain-specific adaptation procedures. The predictions of these refined models are hence blended using a Mixture of Experts approach, which allows for selecting the most reliable for predicting the new examples. The proposed approach is fully cross-domain and does not necessitate retraining or fine-tuning when encountering new domains, thus streamlining the adaptation process and ensuring scalability across diverse data landscapes. Experiments conducted on 5 real datasets demonstrate the robustness and effectiveness of our proposal.