Heroes, Villains, and Victims, and GPT-3: Automated Extraction of Character Roles Without Training Data

Add code
May 17, 2022
Dominik Stammbach, Maria Antoniak, Elliott Ash
Figure 1 for Heroes, Villains, and Victims, and GPT-3: Automated Extraction of Character Roles Without Training Data
Figure 2 for Heroes, Villains, and Victims, and GPT-3: Automated Extraction of Character Roles Without Training Data
Figure 3 for Heroes, Villains, and Victims, and GPT-3: Automated Extraction of Character Roles Without Training Data
Figure 4 for Heroes, Villains, and Victims, and GPT-3: Automated Extraction of Character Roles Without Training Data

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon

This paper shows how to use large-scale pre-trained language models to extract character roles from narrative texts without training data. Queried with a zero-shot question-answering prompt, GPT-3 can identify the hero, villain, and victim in diverse domains: newspaper articles, movie plot summaries, and political speeches.

View paper onarxiv icon

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon