Coarse2Fine: Fine-grained Text Classification on Coarsely-grained Annotated Data

TL;DR

This paper introduces a coarse-to-fine classification framework that leverages pre-trained generative models and weak supervision to perform fine-grained classification on coarsely annotated data, reducing the need for detailed labels.

Contribution

It proposes a novel label-conditioned finetuning approach and a regularization method to improve fine-grained classification using only coarse labels and generative models.

Findings

Outperforms state-of-the-art zero-shot classifiers on real datasets.

Effectively leverages coarse labels for fine-grained classification.

Demonstrates robustness across different datasets and settings.

Abstract

Existing text classification methods mainly focus on a fixed label set, whereas many real-world applications require extending to new fine-grained classes as the number of samples per label increases. To accommodate such requirements, we introduce a new problem called coarse-to-fine grained classification, which aims to perform fine-grained classification on coarsely annotated data. Instead of asking for new fine-grained human annotations, we opt to leverage label surface names as the only human guidance and weave in rich pre-trained generative language models into the iterative weak supervision strategy. Specifically, we first propose a label-conditioned finetuning formulation to attune these generators for our task. Furthermore, we devise a regularization objective based on the coarse-fine label constraints derived from our problem setting, giving us even further improvements over the prior formulation. Our framework uses the fine-tuned generative models to sample pseudo-training data for training the classifier, and bootstraps on real unlabeled data for model refinement. Extensive experiments and case studies on two real-world datasets demonstrate superior performance over SOTA zero-shot classification baselines.