Abstract
Self-training methods have been explored in recent years and have exhibited great performance in improving semi-supervised learning. This work presents a simple instance-adaptive self-training method (SAT) for semi-supervised text classification. SAT first generates two augmented views for each unlabeled data, and then trains a meta learner to automatically identify the relative strength of augmentations based on the similarity between the original view and the augmented views. The weakly-augmented view is fed to the model to produce a pseudo-label and the strongly-augmented view is used to train the model to predict the same pseudo-label. We conducted extensive experiments and analyses on three text classification datasets and found that with varying sizes of labeled training data, SAT consistently shows competitive performance compared to existing semi-supervised learning methods.- Anthology ID:
- 2022.findings-emnlp.456
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2022
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates
- Editors:
- Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 6141–6146
- Language:
- URL:
- https://aclanthology.org/2022.findings-emnlp.456
- DOI:
- 10.18653/v1/2022.findings-emnlp.456
- Cite (ACL):
- Hui Chen, Wei Han, and Soujanya Poria. 2022. SAT: Improving Semi-Supervised Text Classification with Simple Instance-Adaptive Self-Training. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 6141–6146, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
- Cite (Informal):
- SAT: Improving Semi-Supervised Text Classification with Simple Instance-Adaptive Self-Training (Chen et al., Findings 2022)
- PDF:
- https://preview.aclanthology.org/emnlp22-frontmatter/2022.findings-emnlp.456.pdf