Open access
Sep 2026
Explainable Fake News Detection Using Transformers: A Comparative Study of SHAP and Integrated Gradients with Cross-Dataset Generalization
A complete pipeline for explainable fake news detection from the WELFake dataset, consisting of 72,134 articles from 4 sources, that is reproducible and reveals the wire service dateline tokens "Washington" and "Reuters" are the model's strongest evidence for real news.
Sabiha Shahid, Iftikhar Naseer
· International Journal of Inn... · 0 citations