HonestBait: Headline Generation via Faithful Forward Reference

Anonymous

Abstract: Current methods for generating attractive headlines often learn directly from data, which bases attractiveness on the number of user clicks and views. Although clicks or views do reflect user interest, they can fail to reveal how much interest is raised by the writing style and how much is caused by the event or topic itself. Also, such approaches can lead to harmful hallucinations by over-exaggerating the content, aggravating the spread of false information. In this work, we propose HonestBait, a novel framework for solving these issues from another aspect generating headlines using forward references(FR), a writing technique often used in clickbait. A self-verification process is also included in training to avoid harmful hallucinations. We start with a preliminary user study to understand how FR affects user interest, after which we present PANCO, an innovative dataset containing pairs of fake news with verified news020for attractive but faithful news headline generation. Automatic metrics and human evaluations show our framework yields better results in attractiveness while maintaining high veracity.