Skip to content

Latest commit

 

History

History
38 lines (24 loc) · 787 Bytes

README.md

File metadata and controls

38 lines (24 loc) · 787 Bytes

PL-Mix

This work has been accepted as the long paper ''On Unsupervised Domain Adaptation: Pseudo Label Guided Mixup for Adversarial Prompt Tuning'' in AAAI 2024.

Requirements:

python==3.8.10

numpy==1.24.3

tokenizers==0.9.4

torch==1.10.0

transformers==4.1.1

Run:

You can start PL-Mix directly by running the following code:

bash run_all.sh

Cite:

@article{kong2024plmix,
	title={On Unsupervised Domain Adaptation: Pseudo Label Guided Mixup for Adversarial Prompt Tuning},
	author={Fanshuang Kong, Richong Zhang, Ziqiao Wang and Yongyi Mao},
	booktitle={Proceedings of the AAAI conference on artificial intelligence},
	year={2024},
}

Acknowledgement:

Thank you @KanadeSiina for your support of this project!