AdaSent: Efficient Domain-Adapted Sentence Embeddings for Few-Shot Classification

Yongxin Huang, Kexin Wang, Sourav Dutta, Raj Patel, Goran Glavaš, Iryna Gurevych


Abstract
Recent work has found that few-shot sentence classification based on pre-trained Sentence Encoders (SEs) is efficient, robust, and effective. In this work, we investigate strategies for domain-specialization in the context of few-shot sentence classification with SEs. We first establish that unsupervised Domain-Adaptive Pre-Training (DAPT) of a base Pre-trained Language Model (PLM) (i.e., not an SE) substantially improves the accuracy of few-shot sentence classification by up to 8.4 points. However, applying DAPT on SEs, on the one hand, disrupts the effects of their (general-domain) Sentence Embedding Pre-Training (SEPT). On the other hand, applying general-domain SEPT on top of a domain-adapted base PLM (i.e., after DAPT) is effective but inefficient, since the computationally expensive SEPT needs to be executed on top of a DAPT-ed PLM of each domain. As a solution, we propose AdaSent, which decouples SEPT from DAPT by training a SEPT adapter on the base PLM. The adapter can be inserted into DAPT-ed PLMs from any domain. We demonstrate AdaSent’s effectiveness in extensive experiments on 17 different few-shot sentence classification datasets. AdaSent matches or surpasses the performance of full SEPT on DAPT-ed PLM, while substantially reducing the training costs. The code for AdaSent is available.
Anthology ID:
2023.emnlp-main.208
Volume:
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3420–3434
Language:
URL:
https://aclanthology.org/2023.emnlp-main.208
DOI:
10.18653/v1/2023.emnlp-main.208
Bibkey:
Cite (ACL):
Yongxin Huang, Kexin Wang, Sourav Dutta, Raj Patel, Goran Glavaš, and Iryna Gurevych. 2023. AdaSent: Efficient Domain-Adapted Sentence Embeddings for Few-Shot Classification. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 3420–3434, Singapore. Association for Computational Linguistics.
Cite (Informal):
AdaSent: Efficient Domain-Adapted Sentence Embeddings for Few-Shot Classification (Huang et al., EMNLP 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.emnlp-main.208.pdf
Video:
 https://aclanthology.org/2023.emnlp-main.208.mp4