Solving text classification in a weakly supervised manner is important for real-world applications where human annotations are scarce. In this paper, we propose to query a masked language model with cloze style prompts to obtain supervision signals. We design a prompt which combines the document itself and "this article is talking about [MASK]." A masked language model can generate words for the [MASK] token. The generated words which summarize the content of a document can be utilized as supervision signals. We propose a latent variable model to learn a word distribution learner which associates generated words to pre-defined categories and a document classifier simultaneously without using any annotated data. Evaluation on three datasets, AGNews, 20Newsgroups, and UCINews, shows that our method can outperform baselines by 2%, 4%, and 3%.
@article{arxiv.2205.06604,
title = {Weakly Supervised Text Classification using Supervision Signals from a Language Model},
author = {Ziqian Zeng and Weimin Ni and Tianqing Fang and Xiang Li and Xinran Zhao and Yangqiu Song},
journal= {arXiv preprint arXiv:2205.06604},
year = {2022}
}