Methods for addressing class imbalance in deep learning-based natural language processing

Figure 1: Modern Transformer-based Natural Language Processing (NLP) methods still struggle with class imbalance: class-wise performance (second row, each dot represents one class) decreases with class frequency in training data (first row) for a variety of NLP tasks. Datasets/scores: TACRED, Zhou and Chen [2021], USPTO, Pujari et al. [2021], PDTB, Shi and Demberg [2019], UD-EWT, […]

articles chen data datasets deep dive deep learning language language processing natural natural language natural language processing nlp performance processing training training data transformer uspto