Cross-Domain Sentiment Classification with Contrastive Learning and Mutual Information Maximization

TL;DR

This paper introduces CLIM, a novel contrastive learning framework with mutual information maximization, to improve cross-domain sentiment classification, achieving state-of-the-art results by enhancing feature robustness and class separation.

Contribution

First to apply contrastive learning with mutual information maximization for NLP cross-domain tasks, enhancing feature support and classifier robustness.

Findings

Achieved state-of-the-art results on Amazon-review and airlines datasets.

Demonstrated improved robustness and class separation across domains.

Validated effectiveness of CLIM in cross-domain sentiment classification.

Abstract

Contrastive learning (CL) has been successful as a powerful representation learning method. In this work we propose CLIM: Contrastive Learning with mutual Information Maximization, to explore the potential of CL on cross-domain sentiment classification. To the best of our knowledge, CLIM is the first to adopt contrastive learning for natural language processing (NLP) tasks across domains. Due to scarcity of labels on the target domain, we introduce mutual information maximization (MIM) apart from CL to exploit the features that best support the final prediction. Furthermore, MIM is able to maintain a relatively balanced distribution of the model's prediction, and enlarges the margin between classes on the target domain. The larger margin increases our model's robustness and enables the same classifier to be optimal across domains. Consequently, we achieve new state-of-the-art results on the Amazon-review dataset as well as the airlines dataset, showing the efficacy of our proposed method CLIM.