ExTax unifies 6 persuasive, 5 emotional, and 6 narrative categories into a 17-dimensional space, elicits labels from multiple LLMs, reconciles them with entropy-driven smoothing, and fuses them via heterogeneous attention to reach 0.8456 Macro F1 on five disinformation benchmarks while staying robus
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ExTax: Explainable Disinformation Detection via Persuasion, Emotion, and Narrative Role Taxonomies
ExTax unifies 6 persuasive, 5 emotional, and 6 narrative categories into a 17-dimensional space, elicits labels from multiple LLMs, reconciles them with entropy-driven smoothing, and fuses them via heterogeneous attention to reach 0.8456 Macro F1 on five disinformation benchmarks while staying robus