Zero-shot text classification via self-supervised tuning

• Main Authors: Liu, Chaoqun, Zhang, Wenxuan, Chen, Guizhen, Wu, Xiaobao, Luu, Anh Tuan, Chang, Chip Hong, Bing, Lidong
• Other Authors: Interdisciplinary Graduate School (IGS)
• Format: Conference or Workshop Item
• Language: English
• Published: 2023
• Subjects: Computer Science - Computation and Language; Computer Science - Artificial Intelligence; Engineering::Computer science and engineering::Computing methodologies::Document and text processing; Zero-Shot Text Classification; Self-Supervised Tuning
• Online Access: https://hdl.handle.net/10356/168505; https://2023.aclweb.org/
• Institution: Nanyang Technological University

Existing solutions to zero-shot text classification either conduct prompting with pre-trained language models, which is sensitive to the choices of templates, or rely on large-scale annotated data of relevant tasks for meta-tuning. In this work, we propose a new paradigm based on self-supervised learning to solve zero-shot text classification tasks by tuning the language models with unlabeled data, called self-supervised tuning. By exploring the inherent structure of free texts, we propose a new learning objective called first sentence prediction to bridge the gap between unlabeled data and text classification tasks. After tuning the model to learn to predict the first sentence in a paragraph based on the rest, the model is able to conduct zero-shot inference on unseen tasks such as topic classification and sentiment analysis. Experimental results show that our model outperforms the state-of-the-art baselines on 7 out of 10 tasks. Moreover, the analysis reveals that our model is less sensitive to the prompt design. Our code and pre-trained models are publicly available at https://github.com/DAMO-NLP-SG/SSTuning .