XPROMPT: Exploring the Extreme of Prompt Tuning

Fang Ma, Chen Zhang, Lei Ren, Jingang Wang*, Qifan Wang, Wei Wu, Xiaojun Quan, Dawei Song*

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

14 Citations (Scopus)

Abstract

Prompt tuning learns soft prompts to condition the frozen Pre-trained Language Models (PLMs) for performing downstream tasks in a parameter-efficient manner. While prompt tuning has gradually reached the performance level of fine-tuning as the model scale increases, there is still a large performance gap between prompt tuning and fine-tuning for models of moderate and small scales (typically less than 11B parameters). In this paper, we empirically show that the trained prompt tokens can have a negative impact on a downstream task and thus degrade its performance. To bridge the gap, we propose a novel PROMPT tuning model with an eXtremely small scale (XPROMPT) under the regime of lottery tickets hypothesis. Specifically, XPROMPT eliminates the negative prompt tokens at different granularity levels through a hierarchical structured pruning, yielding a more parameter-efficient prompt yet with a competitive performance. Comprehensive experiments are carried out on the SuperGLUE tasks, and the results indicate that XPROMPT is able to close the performance gap at smaller model scales.

Original languageEnglish
Pages11033-11047
Number of pages15
Publication statusPublished - 2022
Event2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022 - Abu Dhabi, United Arab Emirates
Duration: 7 Dec 202211 Dec 2022

Conference

Conference2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022
Country/TerritoryUnited Arab Emirates
CityAbu Dhabi
Period7/12/2211/12/22

Fingerprint

Dive into the research topics of 'XPROMPT: Exploring the Extreme of Prompt Tuning'. Together they form a unique fingerprint.

Cite this

Ma, F., Zhang, C., Ren, L., Wang, J., Wang, Q., Wu, W., Quan, X., & Song, D. (2022). XPROMPT: Exploring the Extreme of Prompt Tuning. 11033-11047. Paper presented at 2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022, Abu Dhabi, United Arab Emirates.