korawat-tanwisuth / POUF

Official implementation for POUF (Prompt-oriented unsupervised fine-tuning for large pre-trained models)

• Introduction
• Instruction
• Citation

Introduction

POUF is an unsupervised fine-tuning framework for adapting prompt-based zero-shot models on the target data.

Instruction

Please refer to the subfolders pouf_lvm for image experiments and pouf_mlm for text experiments.

Citation

If you find this code useful for your research, please cite our paper.

@article{tanwisuth2023pouf,
  title={POUF: Prompt-oriented unsupervised fine-tuning for large pre-trained models},
  author={Tanwisuth, Korawat and Zhang, Shujian and Zheng, Huangjie and He, Pengcheng and Zhou, Mingyuan},
  journal={arXiv preprint arXiv:2305.00350},
  year={2023}
}