Through prompting, large-scale pre-trained models have become more expressive and powerful, gaining significant attention in recent years. Though these big models have zero-shot capabilities, in general, labeled data are still required to adapt them to downstream tasks. To overcome this critical limitation, we propose an unsupervised fine-tuning framework to directly fine-tune the model or prompt on the unlabeled target data. We demonstrate how to apply our method to both language-augmented vision and masked-language models by aligning the discrete distributions extracted from the prompts and target data. To verify our approach's applicability, we conduct extensive experiments on image classification, sentiment analysis, and natural language inference tasks. Across 13 image-related tasks and 15 language-related ones, the proposed approach achieves consistent improvements over the baselines.
@article{arxiv.2305.00350,
title = {POUF: Prompt-oriented unsupervised fine-tuning for large pre-trained models},
author = {Korawat Tanwisuth and Shujian Zhang and Huangjie Zheng and Pengcheng He and Mingyuan Zhou},
journal= {arXiv preprint arXiv:2305.00350},
year = {2023}
}
Comments
ICML 2023; PyTorch code is available at https://github.com/korawat-tanwisuth/POUF