Abstract
Few-shot text classification has attracted great interest in both academia and industry due to the lack of labeled data in many fields. Different from general text classification (e.g., topic classification), few-shot sentiment classification is more challenging because the semantic distances among the classes are more subtle. For instance, the semantic distances between the sentiment labels in a positive or negative polarity (e.g., ``love" and ``joy", ``remorse" and ``sadness") are close, while the distances are large for the sentiment labels in two opposite polarities (e.g., ``love" and ``sadness"). To address this problem, we propose a Soft Contrastive learning-based Prompt (\texttt{SCP}) model for few-shot sentiment analysis. First, we design a sentiment-aware chain of thought prompt module to guide the model to predict the sentiment from coarse grain to fine grain via a series of intermediate reasoning steps. Then, we propose a soft contrastive learning algorithm to take the correlation of the labels into account. A series of experiments on several sentiment analysis datasets show the great advantages of \texttt{SCP} by comparing it with SOTA baselines (e.g., ChatGPT).
Abstract (translated)
由于许多领域缺乏足够的标记数据，稀疏shot文本分类在学术界和产业界都引起了极大的兴趣。与通用文本分类（如主题分类）不同，稀疏 shot 情感分类更具挑战性，因为类别之间的语义距离更加微妙。例如，在正面或负面极性（例如，“爱”和“快乐”，“悲伤”和“忧愁”）中，情感标签之间的语义距离很近，而两个相反极性（例如，“爱”和“悲伤”）中情感标签之间的距离较大。为解决这个问题，我们提出了一个基于软对比学习的学习提示（SCP）模型来进行稀疏 shot 情感分析。首先，我们设计了一个情感意识的思想提示模块，通过一系列中间推理步骤将模型从粗粒度到细粒度预测情感。然后，我们提出了一种软对比学习算法，以考虑标签的相关性。在多个情感分析数据集上的实验表明，与最先进的基准模型（如 ChatGPT）相比，SCP 具有显著的优势。
URL
https://arxiv.org/abs/2312.10479
