VIBE: Topic-Driven Temporal Adaptation for Twitter Classification
This addresses temporal adaptation for Twitter classification, offering a novel approach to handle noisy, evolving data, though it is incremental in improving over existing methods.
The paper tackles the problem of deteriorating text classification performance on evolving social media by proposing VIBE, a model that uses topic-driven temporal adaptation, which significantly outperforms previous state-of-the-art methods using only 3% of data.
Language features are evolving in real-world social media, resulting in the deteriorating performance of text classification in dynamics. To address this challenge, we study temporal adaptation, where models trained on past data are tested in the future. Most prior work focused on continued pretraining or knowledge updating, which may compromise their performance on noisy social media data. To tackle this issue, we reflect feature change via modeling latent topic evolution and propose a novel model, VIBE: Variational Information Bottleneck for Evolutions. Concretely, we first employ two Information Bottleneck (IB) regularizers to distinguish past and future topics. Then, the distinguished topics work as adaptive features via multi-task training with timestamp and class label prediction. In adaptive learning, VIBE utilizes retrieved unlabeled data from online streams created posterior to training data time. Substantial Twitter experiments on three classification tasks show that our model, with only 3% of data, significantly outperforms previous state-of-the-art continued-pretraining methods.