A Simple Yet Strong Domain-Agnostic De-bias Method for Zero-Shot Sentiment Classification

Zero-shot prompt-based learning has made much progress in sentiment analysis, and considerable effort has been dedicated to designing high-performing prompt templates. However, two problems exist; First, large language models are often biased to their pre-training data, leading to poor performance in prompt templates that models have rarely seen. Second, in order to adapt to different domains, redesigning prompt templates is usually required, which is time-consuming and inefficient. To remedy both shortcomings, we propose a simple yet strong data construction method to debias a given prompt template, yielding a large performance improvement in sentiment analysis tasks across different domains, pre-trained language models, and prompt templates. Also, we demonstrate the advantage of using domainagnostic generic responses over the in-domain ground-truth data. We release the code here 1 .