MALicious INTent Dataset and Inoculating LLMs for Enhanced Disinformation Detection

2026-03-15

AI Summary

This study addresses a critical gap in current misinformation detection research—the lack of explicit modeling and annotated data for malicious intent, which hinders the identification of deliberately deceptive content. To bridge this gap, the authors introduce MALINT, the first English-language corpus annotated by professional fact-checkers for malicious intent, grounded in psychological inoculation theory. They further propose an intent-augmented reasoning mechanism that integrates malicious intent analysis into the zero-shot inference process of large language models, such as Llama 3.3. Experimental results demonstrate that this approach significantly enhances the misinformation detection performance of five different large language models across six benchmark datasets. The MALINT dataset has been publicly released to advance research in intent-aware misinformation detection.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Application Domains: Misinformation & Fake News

Application Category

• Responsible Web: Mitigating misinformation and disinformation
• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

The intentional creation and spread of disinformation poses a significant threat to public discourse. However, existing English datasets and research rarely address the intentionality behind the disinformation. This work presents MALINT, the first human-annotated English corpus developed in collaboration with expert fact-checkers to capture disinformation and its malicious intent. We utilize our novel corpus to benchmark 12 language models, including small language models (SLMs) such as BERT and large language models (LLMs) like Llama 3.3, on binary and multilabel intent classification tasks. Moreover, inspired by inoculation theory from psychology and communication studies, we investigate whether incorporating knowledge of malicious intent can improve disinformation detection. To this end, we propose intent-based inoculation, an intent-augmented reasoning for LLMs that integrates intent analysis to mitigate the persuasive impact of disinformation. Analysis on six disinformation datasets, five LLMs, and seven languages shows that intent-augmented reasoning improves zero-shot disinformation detection. To support research in intent-aware disinformation detection, we release the MALINT dataset with annotations from each annotation step.

Problem

• disinformation
• malicious intent
• intent detection
• fact-checking
• language models

Innovation

• malicious intent
• disinformation detection
• intent-based inoculation
• MALINT dataset
• intent-augmented reasoning