Adaptive Prompt Learning-based Few-Shot Sentiment Analysis

TL;DR

This paper introduces an adaptive prompt learning strategy for few-shot sentiment analysis that dynamically constructs prompts using a seq2seq-attention model, improving performance and generalization across datasets.

Contribution

It proposes a novel adaptive prompt construction method leveraging seq2seq-attention, enhancing prompt quality and cross-domain applicability in few-shot sentiment analysis.

Findings

Outperforms state-of-the-art baselines on FewCLUE datasets

Effectively constructs prompts regardless of hand-crafted prompt quality

Improves model performance in data-scarce sentiment analysis scenarios

Abstract

In the field of natural language processing, sentiment analysis via deep learning has a excellent performance by using large labeled datasets. Meanwhile, labeled data are insufficient in many sentiment analysis, and obtaining these data is time-consuming and laborious. Prompt learning devotes to resolving the data deficiency by reformulating downstream tasks with the help of prompt. In this way, the appropriate prompt is very important for the performance of the model. This paper proposes an adaptive prompting(AP) construction strategy using seq2seq-attention structure to acquire the semantic information of the input sequence. Then dynamically construct adaptive prompt which can not only improve the quality of the prompt, but also can effectively generalize to other fields by pre-trained prompt which is constructed by existing public labeled data. The experimental results on FewCLUE datasets demonstrate that the proposed method AP can effectively construct appropriate adaptive prompt regardless of the quality of hand-crafted prompt and outperform the state-of-the-art baselines.