Papers › FNR: a similarity and transformer-based approach to detect multi-modal fake news in...
FNR: a similarity and transformer-based approach to detect multi-modal fake news in social media
Faeze Ghorbanpour, Maryam Ramezani, Mohammad Amin Fazli, Hamid R. Rabiee
Many people today get their news from social media. It is possible to propagate news using textual, visual, or multi-modal information. The popularity of social networks and their wide use by people make them attractive platforms for spreading fake news. Detecting fake news is essential to preventing its spread. Fake news can be a false article or a genuine article with misleading visual information. Adding an actual image to trustworthy unrelated news can also create a fake news story. In this paper, we propose a novel and efficient similarity and transformer-based detection algorithm called Fake News Revealer (FNR), which uses text and images of news to detect fake news. The algorithm uses contrastive loss to consider text and image relations and transformer models to extract contextual and semantic features. According to experiments on two public social media news data sets, the FNR algorithm competes with conventional methods and state-of-the-art fake news detection algorithms by adding a novel mechanism without adding extra parameters or weights.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
No leaderboard rows for this paper in the archive.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections