Agentic Detection of Online Conspiracies

In 30 Sekunden

Was passiert ist

Researchers developed an agentic framework that detects conspiracy theories on social media by reasoning about speaker intent using social context rather than text alone.

Warum es zählt

Matters for content moderators, researchers studying misinformation, and engineers building systems to identify coordinated inauthentic behavior on platforms.

Achtung

Tested only on Hebrew tweets; unclear how well the approach generalizes to other languages, platforms, or types of conspiracy discourse beyond the dataset period.