« Back to Glossary Index

Zero-shot extraction refers to the ability of an AI model to extract relevant data from text without prior training on specific examples or annotations. The model can perform this task based on its understanding of language and the patterns it has learned from general data, enabling it to handle new or unseen data effectively without the need for domain-specific training.