dhfbk/faina

Fine-grained Fallacy Detection with Human Label Variation (NAACL 2025)

14
/ 100
Experimental

This project offers a specialized dataset and models for identifying different kinds of logical fallacies in social media posts. It takes Italian social media text as input and helps identify specific fallacies, even when human experts might disagree. Researchers and analysts focused on understanding and combating misinformation, particularly in Italian online discussions, would use this.

No commits in the last 6 months.

Use this if you need to analyze Italian social media text to automatically detect and categorize logical fallacies, especially if you're interested in how human experts might interpret the same text differently.

Not ideal if your primary goal is to analyze English text or if you need a simple 'true/false' fallacy detection without fine-grained categorization or consideration of annotator disagreement.

misinformation-detection social-media-analysis argumentation-analysis online-discourse text-analysis
No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 4 / 25
Maturity 8 / 25
Community 0 / 25

How are scores calculated?

Stars

5

Forks

Language

Python

License

Last pushed

Sep 05, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/dhfbk/faina"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.