ClickGuard: Detecting and Spoiling Clickbait News with Informativeness Measures and Large Language Models

Short summary

ClickGuard is a browser extension that detects clickbait using a hybrid ML architecture combining transformer embeddings, linguistic features, and a custom 'baitness' score, achieving 91% F1 on an open dataset. It warns users before and after accessing suspicious articles with a percentage score and explanation. The tool also generates a one-to-two-sentence spoiler summarizing the article's actual content.

• •XGBoost-based clickbait detector achieves 91% F1 using transformer embeddings and custom linguistic features
• •Browser extension warns users with percentage scores and metric-based explanations before and after article access
• •Includes clickbait spoiler feature providing concise article summaries

Generated with AI, which can make mistakes.