BanClickThumb: A Multimodal Dataset and Transformer Fusion Benchmarks for Clickbait Detection in Bengali YouTube Videos
Computer Vision and Pattern Recognition
Summary
The authors created a new dataset called BanClickThumb to help detect clickbait in Bengali YouTube videos by looking at both the video titles and thumbnails. They tested different methods using just the text, just the images, and a combination of both, finding that combining them worked best for spotting clickbait. Their best model, BanClickFusionFormer, used advanced language and image models together to reach 84% accuracy. They also noticed that some challenges remain, like tricky language and cultural slang. This work provides a useful starting point for detecting misleading content in low-resource languages.
Authors
Md. Ariful Islam, Md Tanvirul Islam, Md. Maruf Hossain Miru, Md Khalid Syfullah
Abstract
Clickbait, where video titles and thumbnails exaggerate or misrepresent content, reduces user trust, wastes attention, and promotes misinformation on video-sharing platforms. Detecting Bengali clickbait remains challenging because publicly available multimodal datasets are limited. To address this gap, we introduce BanClickThumb, a curated dataset of 7,147 Bengali YouTube thumbnail-title pairs from five content domains, annotated by ten annotators with high agreement (Cohen's Kappa: 0.83-0.93). Using this dataset, we benchmark text-only, image-only, and multimodal approaches. Among unimodal models, BanClickTextFormer (XLM-RoBERTa) achieves 0.82 accuracy, while BanClickImageFormer (SwiftFormer) reaches 0.68. Our proposed multimodal model, BanClickFusionFormer, combines ViT and XLM-RoBERTa through intermediate fusion and achieves the best accuracy of 0.84. Error analysis shows that dense thumbnail text, figurative language, and culturally specific slang remain challenging. Our findings demonstrate the effectiveness of multimodal fusion for Bengali clickbait detection and provide a publicly available benchmark to support future research on low-resource multimodal content analysis.