Contrastive Learning for Fake News
Collection
Datasets and model for training models with contrastive loss. Domain/task: Fake news / hoax classification.
•
15 items
•
Updated
Data source: nlp-brin-id/triplets-all, but excluding 'Fact' as triplets.