Researchers have developed a new model for detecting rumors in social media posts that combine images and text. This model enhances detection by incorporating external factual evidence and analyzing forgery features within the content. It utilizes a ResNet34 visual encoder and a BERT text encoder, augmented by a forgery feature module that processes frequency domain data through Fourier transformation. The system also employs BLIP for generating concise, image-faithful descriptions to improve semantic alignment between text and images, and a gated adaptive feature scaling fusion mechanism for dynamic multimodal integration. Experiments on Weibo and Twitter datasets show improved performance in accuracy, recall, and F1 score compared to existing methods. AI
IMPACT This research could lead to more robust systems for identifying misinformation on social media platforms.
RANK_REASON Academic paper detailing a new model for rumor detection. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
