ExTax: Explainable Disinformation Detection via Persuasion, Emotion, and Narrative Role Taxonomies
arXiv:2605.27045v1 Announce Type: new Abstract: The democratization of LLMs has accelerated the generation and circulation of highly fluent disinformation, making traditional syntax-semantic verification increasingly insufficient. Such deception rarely relies solely on surface-level falsity; instead, it often combines persuasive rhetoric, emotional manipulation, and narrative role construction to influence readers' interpretations through multiple cognitive pathways. However, existing detectors typically emphasize isolated signals -- such as syntax, external knowledge, persuasion, or affective
The rapid democratization of sophisticated LLMs has created an urgent need for more advanced disinformation detection methods that go beyond surface-level analysis, prompting this novel approach using persuasion, emotion, and narrative taxonomies.
This research is crucial for any strategic reader concerned with information integrity and societal stability, as it directly addresses the escalating challenge of AI-generated disinformation that undermines public discourse and decision-making.
Disinformation detection is beginning to move beyond syntax and isolated signals, incorporating deeper linguistic and psychological analysis to identify manipulative content, making it harder for advanced AI to generate undetectable falsehoods.
- · AI ethics researchers
- · Social media platforms
- · Journalistic integrity organizations
- · Democratic institutions
- · Disinformation actors
- · Malicious LLM users
- · Platforms with weak content moderation
Improved disinformation detection tools become available, enhancing the ability to identify AI-generated manipulative content.
Public trust in information sources could gradually increase as advanced detection methods mitigate the spread of sophisticated falsehoods.
The development of 'disinformation-resistant' LLMs might emerge, where models are inherently designed to avoid generating manipulative narratives, leading to a new paradigm in AI safety and content creation.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL