Multi-Modal Semantic Inconsistency Detection in Social Media News Posts

TL;DR

This paper presents a multi-modal framework for detecting semantic mismatches in social media news posts, combining text, audio, and video analysis to improve accuracy over uni-modal methods.

Contribution

The paper introduces a novel multi-modal fusion architecture and a new dataset for detecting semantic inconsistencies in social media videos and captions.

Findings

Achieves 60.5% accuracy in mismatch detection

Fusion across multiple modalities improves performance

A new dataset of 4,000 Facebook news posts was curated

Abstract

As computer-generated content and deepfakes make steady improvements, semantic approaches to multimedia forensics will become more important. In this paper, we introduce a novel classification architecture for identifying semantic inconsistencies between video appearance and text caption in social media news posts. We develop a multi-modal fusion framework to identify mismatches between videos and captions in social media posts by leveraging an ensemble method based on textual analysis of the caption, automatic audio transcription, semantic video analysis, object detection, named entity consistency, and facial verification. To train and test our approach, we curate a new video-based dataset of 4,000 real-world Facebook news posts for analysis. Our multi-modal approach achieves 60.5% classification accuracy on random mismatches between caption and appearance, compared to accuracy below 50% for uni-modal models. Further ablation studies confirm the necessity of fusion across modalities for correctly identifying semantic inconsistencies.