What type of study is this?

This is a Quantitative Study study.

October 13, 2025Open Access

PolyTruth: Multilingual Disinformation Detection using Transformer-Based Language Models

View Full Paper

Ask AI

Bookmark

Authors

Systematic comparison of five multilingual transformer models reveals varied effectiveness in disinformation accuracy.

RemBERT showed superior accuracy in detecting disinformation, especially in low-resource languages, while others lagged.
The PolyTruth Disinfo Corpus includes 60,486 statement pairs to assess multilingual disinformation detection.
This analysis highlights the importance of multilingual capabilities in AI for effective disinformation classification.
Findings point to both the potential and limitations of current AI systems in tackling disinformation across languages.

Gouliev et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark