SEER: Semantic Enhancement and Emotional Reasoning Network for Multimodal Fake News Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Peican, Jing, Yubo, Cheng, Le, Chen, Bin, Cui, Xiaodong, Wu, Lianwei, Tang, Keke
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909694680891392
author Zhu, Peican
Jing, Yubo
Cheng, Le
Chen, Bin
Cui, Xiaodong
Wu, Lianwei
Tang, Keke
author_facet Zhu, Peican
Jing, Yubo
Cheng, Le
Chen, Bin
Cui, Xiaodong
Wu, Lianwei
Tang, Keke
contents Previous studies on multimodal fake news detection mainly focus on the alignment and integration of cross-modal features, as well as the application of text-image consistency. However, they overlook the semantic enhancement effects of large multimodal models and pay little attention to the emotional features of news. In addition, people find that fake news is more inclined to contain negative emotions than real ones. Therefore, we propose a novel Semantic Enhancement and Emotional Reasoning (SEER) Network for multimodal fake news detection. We generate summarized captions for image semantic understanding and utilize the products of large multimodal models for semantic enhancement. Inspired by the perceived relationship between news authenticity and emotional tendencies, we propose an expert emotional reasoning module that simulates real-life scenarios to optimize emotional features and infer the authenticity of news. Extensive experiments on two real-world datasets demonstrate the superiority of our SEER over state-of-the-art baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2507_13415
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SEER: Semantic Enhancement and Emotional Reasoning Network for Multimodal Fake News Detection
Zhu, Peican
Jing, Yubo
Cheng, Le
Chen, Bin
Cui, Xiaodong
Wu, Lianwei
Tang, Keke
Multimedia
Artificial Intelligence
Previous studies on multimodal fake news detection mainly focus on the alignment and integration of cross-modal features, as well as the application of text-image consistency. However, they overlook the semantic enhancement effects of large multimodal models and pay little attention to the emotional features of news. In addition, people find that fake news is more inclined to contain negative emotions than real ones. Therefore, we propose a novel Semantic Enhancement and Emotional Reasoning (SEER) Network for multimodal fake news detection. We generate summarized captions for image semantic understanding and utilize the products of large multimodal models for semantic enhancement. Inspired by the perceived relationship between news authenticity and emotional tendencies, we propose an expert emotional reasoning module that simulates real-life scenarios to optimize emotional features and infer the authenticity of news. Extensive experiments on two real-world datasets demonstrate the superiority of our SEER over state-of-the-art baselines.
title SEER: Semantic Enhancement and Emotional Reasoning Network for Multimodal Fake News Detection
topic Multimedia
Artificial Intelligence
url https://arxiv.org/abs/2507.13415