FIELD NOTES TOPIC
early warning system
1 note, newest first.
Contextualized Early Detection of Online Firestorms: A Sequential LLM-Based Approach
This research paper proposes an LLM (Language Model) based system to detect online firestorms, which are rapid escalations of negative user-generated content, on Reddit. The system operates in two modes: retrospective…
