Publication Details

Abstract

This paper explores enhancing machine learning models' ability to generate accurate results with minimal data through few-shot and zero-shot learning techniques. These techniques leverage existing results and approaches to achieve high accuracy when training datasets are limited. The study focuses on model adaptation and generalization, aiming to provide a comprehensive understanding of how these methods can be effectively applied to improve performance in data-scarce scenarios. The field of machine learning has continually evolved to address the challenges posed by limited data availability, leading to the development of few-shot and zero-shot learning paradigms (Parmar et al., 2025).