Propasafe-Hybrid detects propaganda sentences with clear explanations
Propasafe-Hybrid: A Text-Based Hybrid Propaganda Detection Tool
Computation and LanguageComputers and Society
Summary
Online news and social media often contain misleading propaganda that people quickly read without careful checks. The authors created Propasafe-Hybrid, a tool that spots propaganda in individual sentences and tells you exactly which techniques are used and why. It mixes fast automatic checks with smarter but costly large language models to balance speed and detail. This way, users get clear highlights and short explanations about misleading content without wasting computing power.
What this means in practice
- •For news moderation teams: Flag and label propaganda sentences in news articles with clear explanations to aid human review.
- •For social media platform operators: Automatically detect and explain propaganda in posts to improve content quality and user trust.
Authors
Thomas Kimmeth, Avijit Roy, Vivek Sharma
Abstract
Propagandistic content increasingly circulates through online news and social media, where readers often encounter it with limited scrutiny, highlighting the need for reliable and fine-grained detection. This paper introduces Propasafe-Hybrid, a sentence-level system that integrates a fine-tuned transformer classifier with LLM-based technique classification to identify, label, and explain specific propaganda strategies. The pipeline generates actionable outputs, including highlighted sentences, technique assignments, and concise rationales, so users can immediately understand why a sentence was flagged and how each label was determined. To control inference cost, Propasafe-Hybrid employs a cost-aware pre-filtering stage that forwards only high-likelihood sentences to LLMs, reducing token usage while preserving the underlying decision logic. Together, these design choices enhance the explainability, efficiency, and practical usability of sentence-level propaganda detection in real-world news environments.