Interlock-Free Multi-Aspect Rationalization for Text Classification
It addresses a specific issue in generating multiple rationales for text classification, which is incremental but important for interpretability in NLP.
The paper tackles the interlocking problem in multi-aspect rationalization for text classification by proposing a multi-stage training method with a self-supervised contrastive loss, resulting in significantly improved rationalization performance on the beer review dataset.
Explanation is important for text classification tasks. One prevalent type of explanation is rationales, which are text snippets of input text that suffice to yield the prediction and are meaningful to humans. A lot of research on rationalization has been based on the selective rationalization framework, which has recently been shown to be problematic due to the interlocking dynamics. In this paper, we show that we address the interlocking problem in the multi-aspect setting, where we aim to generate multiple rationales for multiple outputs. More specifically, we propose a multi-stage training method incorporating an additional self-supervised contrastive loss that helps to generate more semantically diverse rationales. Empirical results on the beer review dataset show that our method improves significantly the rationalization performance.