BanClickThumb: A Multimodal Dataset and Transformer Fusion Benchmarks for Clickbait Detection in Bengali YouTube Videos

2026-07-19

AI Summary

This study addresses the absence of publicly available datasets for multimodal clickbait detection in Bengali by introducing the first annotated dataset comprising 7,147 YouTube video title–thumbnail pairs. The authors propose BanClickFusionFormer, a multimodal architecture that encodes textual input using XLM-RoBERTa and visual input via Vision Transformer (ViT) or SwiftFormer, integrating these representations through an intermediate fusion strategy. Experimental results demonstrate that the proposed model achieves an accuracy of 0.84 on the test set, significantly outperforming unimodal baselines (text-only: 0.82; image-only: 0.68). This work advances multimodal analysis for low-resource languages and contributes to the research community by publicly releasing the curated dataset to facilitate future investigations.

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.

Problem

• clickbait detection
• Bengali YouTube videos
• multimodal dataset
• title-thumbnail misrepresentation
• low-resource content analysis

Innovation

• multimodal fusion
• clickbait detection
• low-resource language
• Transformer
• benchmark dataset