Enhancing Fact-Checking: From Crowdsourced Validation to Integration with Large Language Models

Abstract

This extended abstract presents results from two recent studies aimed at enhancing the practical application and effectiveness of fact-checking systems. The first study demonstrates the practical viability of crowdsourcing in fact-checking through an experimental evaluation using political public statements. The second builds on this foundation by integrating crowdsourced data with large language models and proposing a hybrid system that combines human insights with AI capabilities.