whoisjones/pilener_entropy_splits
Viewer • Updated • 78.8k • 9
Familiarity (Difficult NER training datasets)
Each dataset is split into easy, medium and a difficult split using the familiarity metric. Please see our paper for details.
-
-
whoisjones/pilener_max_splits
Viewer • Updated • 43.7k • 17 -
whoisjones/nuner_entropy_splits
Viewer • Updated • 618k • 16 -
whoisjones/nuner_max_splits
Viewer • Updated • 546k • 27 -
Familiarity: Better Evaluation of Zero-Shot Named Entity Recognition by Quantifying Label Shifts in Synthetic Training Data
Paper • 2412.10121 • Published • 2