Inducer-tuning: Connecting Prefix-tuning and Adapter-tuning

Yifan Chen, Devamanyu Hazarika, Mahdi Namazifar, Yang Liu, Di Jin, Dilek Hakkani-Tur


Abstract
Prefix-tuning, or more generally continuous prompt tuning, has become an essential paradigm of parameter-efficient transfer learning. Using a large pre-trained language model (PLM), prefix-tuning can obtain strong performance by training only a small portion of parameters. In this paper, we propose to understand and further develop prefix-tuning through the kernel lens. Specifically, we make an analogy between prefixes and inducing variables in kernel methods and hypothesize that prefixes serving as inducing variables would improve their overall mechanism. From the kernel estimator perspective, we suggest a new variant of prefix-tuning—inducer-tuning, which shares the exact mechanism as prefix-tuning while leveraging the residual form found in adapter-tuning. This mitigates the initialization issue in prefix-tuning. Through comprehensive empirical experiments on natural language understanding and generation tasks, we demonstrate that inducer-tuning can close the performance gap between prefix-tuning and fine-tuning.
Anthology ID:
2022.emnlp-main.50
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
793–808
Language:
URL:
https://aclanthology.org/2022.emnlp-main.50
DOI:
10.18653/v1/2022.emnlp-main.50
Bibkey:
Cite (ACL):
Yifan Chen, Devamanyu Hazarika, Mahdi Namazifar, Yang Liu, Di Jin, and Dilek Hakkani-Tur. 2022. Inducer-tuning: Connecting Prefix-tuning and Adapter-tuning. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 793–808, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Inducer-tuning: Connecting Prefix-tuning and Adapter-tuning (Chen et al., EMNLP 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-2/2022.emnlp-main.50.pdf