Language Models in the Loop: Incorporating Prompting into Weak Supervision

We propose a new strategy for applying large pre-trained language models to novel tasks when labeled training data is limited. Rather than apply the model in a typical zero-shot or few-shot fashion, we treat the model as the basis for labeling functions in a weak supervision framework. To create a c...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	ACM / IMS journal of data science 2024-06, Vol.1 (2), p.1-30
Hauptverfasser:	Smith, Ryan, Fries, Jason A., Hancock, Braden, Bach, Stephen H.
Format:	Artikel
Sprache:	eng
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!