FakeSpotter: A content and strategy agnostic Viral Misinformation Detection Tool

FakeSpotter：一种与内容和策略无关的病毒式虚假信息检测工具

Abstract: Misinformation detection tools often rely on binary true and false classifications or models trained on historical examples, limiting their usefulness when novel misleading narratives emerge. 摘要： 虚假信息检测工具通常依赖于二元真假分类，或基于历史案例训练的模型，这限制了它们在面对新型误导性叙事时的有效性。

Here, we present FakeSpotter, a content- and strategy-agnostic tool designed to estimate the viral misinformation risk of textual content by measuring structural fingerprints of misinformation rather than directly adjudicating truthfulness. 在此，我们提出了 FakeSpotter，这是一种与内容和策略无关的工具。它旨在通过衡量虚假信息的“结构指纹”来评估文本内容的病毒式传播风险，而不是直接判定其真实性。

FakeSpotter operationalizes a theory-driven framework across linguistic, narrative, logical, and critical-thinking dimensions, using repeated LLM assessments and domain-specific logistic regression classifiers for short and long texts. FakeSpotter 将一个理论驱动的框架付诸实践，涵盖了语言、叙事、逻辑和批判性思维等维度，并针对短文本和长文本使用了重复的大语言模型（LLM）评估以及特定领域的逻辑回归分类器。

In a labelled corpus of 764 texts from social media and FakeNewsNet, FakeSpotter achieved macro F1 scores of 0.788 for short texts and 0.793 for long texts on a held-out test set. 在一个包含 764 条来自社交媒体和 FakeNewsNet 的标注语料库中，FakeSpotter 在留出测试集上取得了优异表现：短文本的宏观 F1 分数为 0.788，长文本为 0.793。

FakeSpotter’s interpretive layer provides explainable outputs through feature-based scores, signal agreement, and a caution index, and can be used for social listening. FakeSpotter 的解释层通过基于特征的评分、信号一致性和警示指数提供可解释的输出，并可用于社交媒体舆情监测。

These findings suggest that identifying the structural fingerprints of misinformation can support early, explainable, and human-supervised assessment of potentially viral misinformation. 这些研究结果表明，识别虚假信息的结构指纹，能够支持对潜在病毒式虚假信息进行早期、可解释且有人工监督的评估。