y0news
← Feed
Back to feed
🧠 AI NeutralImportance 6/10

MALicious INTent Dataset and Inoculating LLMs for Enhanced Disinformation Detection

arXiv – CS AI|Arkadiusz Modzelewski, Witold Sosnowski, Eleni Papadopulos, Elisa Sartori, Tiziano Labruna, Giovanni Da San Martino, Adam Wierzbicki|
🤖AI Summary

Researchers released MALINT, the first human-annotated English dataset for detecting disinformation and its malicious intent, developed with expert fact-checkers. The study benchmarked 12 language models and introduced intent-based inoculation techniques that improved zero-shot disinformation detection across six datasets, five LLMs, and seven languages.

Key Takeaways
  • MALINT is the first English corpus specifically designed to capture both disinformation and the malicious intent behind it.
  • Researchers tested 12 language models including BERT and Llama 3.3 on binary and multilabel intent classification tasks.
  • Intent-based inoculation, inspired by psychology research, integrates intent analysis to improve disinformation detection.
  • The approach showed improvements in zero-shot disinformation detection across multiple languages and datasets.
  • The MALINT dataset has been released publicly to support further research in intent-aware disinformation detection.
Mentioned in AI
Models
LlamaMeta
Read Original →via arXiv – CS AI
Act on this with AI
Stay ahead of the market.
Connect your wallet to an AI agent. It reads balances, proposes swaps and bridges across 15 chains — you keep full control of your keys.
Connect Wallet to AI →How it works
Related Articles