A Dual-branch Learning Model with Gradient-balanced Loss for Long-tailed Multi-label Text Classification

Abstract

Multi-label text classification has a wide range of applications in the real world. However, the data distribution in the real world is often imbalanced, which leads to serious long-tailed problems. For multi-label classification, due to the vast scale of datasets and existence of label co-occurrence, how to effectively improve the prediction accuracy of tail labels without degrading the overall precision becomes an important challenge. To address this issue, we propose A Dual-Branch Learning Model with Gradient-Balanced Loss (DBGB) based on the paradigm of existing pre-trained multi-label classification SOTA models. Our model consists of two main long-tailed module improvements. First, with the shared text representation, the dual-classifier are leveraged to process two kinds of label distributions, the one is the original data distribution and the other is the under-sampling distribution for head labels to strengthen the prediction for tail labels. Second, the proposed gradient-balanced loss can adaptively suppress the negative gradient accumulation problem related to labels, especially tail labels. We perform extensive experiments on three multi-label text classification datasets. The results show that the proposed method achieves competitive performance on overall prediction results compared to the state-of-the-art methods in solving the multi-label classification, with significant improvement on tail-label accuracy.