Fake News Detection and Manipulation Reasoning via Large Vision-Language Models

TL;DR

This paper introduces a new benchmark and a multi-modal vision-language model for fake news detection that emphasizes manipulation reasoning and factual accuracy, outperforming existing models and large language models.

Contribution

It proposes a novel benchmark for manipulation reasoning in fake news detection and develops M-DRUM, a multi-modal model with reasoning capabilities based on LVLMs.

Findings

M-DRUM outperforms state-of-the-art fake news detection models.

The benchmark HFFN emphasizes human-centric and fact-related annotations.

The model effectively detects manipulations and reasons about potential fakes.

Abstract

Fake news becomes a growing threat to information security and public opinion with the rapid sprawl of media manipulation. Therefore, fake news detection attracts widespread attention from academic community. Traditional fake news detection models demonstrate remarkable performance on authenticity binary classification but their ability to reason detailed faked traces based on the news content remains under-explored. Furthermore, due to the lack of external knowledge, the performance of existing methods on fact-related news is questionable, leaving their practical implementation unclear. In this paper, we propose a new multi-media research topic, namely manipulation reasoning. Manipulation reasoning aims to reason manipulations based on news content. To support the research, we introduce a benchmark for fake news detection and manipulation reasoning, referred to as Human-centric and Fact-related Fake News (HFFN). The benchmark highlights the centrality of human and the high factual relevance, with detailed manual annotations. HFFN encompasses four realistic domains with fake news samples generated through three manipulation approaches. Moreover, a Multi-modal news Detection and Reasoning langUage Model (M-DRUM) is presented not only to judge on the authenticity of multi-modal news, but also raise analytical reasoning about potential manipulations. On the feature extraction level, a cross-attention mechanism is employed to extract fine-grained fusion features from multi-modal inputs. On the reasoning level, a large vision-language model (LVLM) serves as the backbone to facilitate fact-related reasoning. A two-stage training framework is deployed to better activate the capacity of identification and reasoning. Comprehensive experiments demonstrate that our model outperforms state-of-the-art (SOTA) fake news detection models and powerful LVLMs like GPT-4 and LLaVA.