Learning to Initialize: Can Meta Learning Improve Cross-task Generalization in Prompt Tuning?

Chengwei Qin, Shafiq Joty, Qian Li, Ruochen Zhao


Abstract
Prompt tuning (PT) which only tunes the embeddings of an additional sequence of tokens per task, keeping the pre-trained language model (PLM) frozen, has shown remarkable performance in few-shot learning. Despite this, PT has been shown to rely heavily on good initialization of the prompt embeddings. In this work, we study meta prompt tuning (MPT) to systematically explore how meta-learning can help improve (if it can) cross-task generalization in PT through learning to initialize the prompt embeddings from other relevant tasks. We empirically analyze a representative set of meta learning algorithms in a wide range of adaptation settings with different source/target task configurations on a large set of few-shot tasks. With extensive experiments and analysis, we demonstrate the effectiveness of MPT. We find the improvement to be significant particularly on classification tasks. For other kinds of tasks such as question answering, we observe that while MPT can outperform PT in most cases, it does not always outperform multi-task learning. We further provide an in-depth analysis from the perspective of task similarity.
Anthology ID:
2023.acl-long.659
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
11802–11832
Language:
URL:
https://aclanthology.org/2023.acl-long.659
DOI:
10.18653/v1/2023.acl-long.659
Bibkey:
Cite (ACL):
Chengwei Qin, Shafiq Joty, Qian Li, and Ruochen Zhao. 2023. Learning to Initialize: Can Meta Learning Improve Cross-task Generalization in Prompt Tuning?. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11802–11832, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Learning to Initialize: Can Meta Learning Improve Cross-task Generalization in Prompt Tuning? (Qin et al., ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-long.659.pdf
Video:
 https://aclanthology.org/2023.acl-long.659.mp4