This paper reviews the intersection of low-shot learning (LSL) and class imbalance (CI) in machine learning. LSL, including few-shot, one-shot, and zero-shot learning, aims to classify unseen classes with limited labeled data, while CI refers to datasets where one class has significantly more instances than others. The authors conduct a survey of over 60 papers published between January 2020 and July 2023, examining methodologies and experimental results. They find that most works report performance at or above state-of-the-art levels and highlight current research gaps, particularly in using LSL techniques for large-scale imbalanced tasks and in combining LSL with large language models or semantic data. The paper is organized into sections covering the background on LSL and CI, methodologies, and findings from the literature review, including specific applications such as image classification, zero-shot image classification, object detection, and other image-data tasks. The authors also discuss the shortcomings and future research directions, emphasizing the need for more conclusive comparisons and novel methods that leverage existing techniques.This paper reviews the intersection of low-shot learning (LSL) and class imbalance (CI) in machine learning. LSL, including few-shot, one-shot, and zero-shot learning, aims to classify unseen classes with limited labeled data, while CI refers to datasets where one class has significantly more instances than others. The authors conduct a survey of over 60 papers published between January 2020 and July 2023, examining methodologies and experimental results. They find that most works report performance at or above state-of-the-art levels and highlight current research gaps, particularly in using LSL techniques for large-scale imbalanced tasks and in combining LSL with large language models or semantic data. The paper is organized into sections covering the background on LSL and CI, methodologies, and findings from the literature review, including specific applications such as image classification, zero-shot image classification, object detection, and other image-data tasks. The authors also discuss the shortcomings and future research directions, emphasizing the need for more conclusive comparisons and novel methods that leverage existing techniques.