Towards Robust Pruning: An Adaptive Knowledge-Retention Pruning Strategy for Language Models

Jianwei Li, Qi Lei, Wei Cheng, Dongkuan Xu


Abstract
The pruning objective has recently extended beyond accuracy and sparsity to robustness in language models. Despite this, existing methods struggle to enhance robustness against adversarial attacks when continually increasing model sparsity and require a retraining process. As humans step into the era of large language models, these issues become increasingly prominent. This paper proposes that the robustness of language models is proportional to the extent of pre-trained knowledge they encompass. Accordingly, we introduce a post-training pruning strategy designed to faithfully replicate the embedding space and feature space of dense language models, aiming to conserve more pre-trained knowledge during the pruning process. In this setup, each layer’s reconstruction error not only originates from itself but also includes cumulative error from preceding layers, followed by an adaptive rectification. Compared to other state-of-art baselines, our approach demonstrates a superior balance between accuracy, sparsity, robustness, and pruning cost with BERT on datasets SST2, IMDB, and AGNews, marking a significant stride towards robust pruning in language models.
Anthology ID:
2023.emnlp-main.79
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:
1229–1247
Language:
URL:
https://aclanthology.org/2023.emnlp-main.79
DOI:
10.18653/v1/2023.emnlp-main.79
Bibkey:
Cite (ACL):
Jianwei Li, Qi Lei, Wei Cheng, and Dongkuan Xu. 2023. Towards Robust Pruning: An Adaptive Knowledge-Retention Pruning Strategy for Language Models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 1229–1247, Singapore. Association for Computational Linguistics.
Cite (Informal):
Towards Robust Pruning: An Adaptive Knowledge-Retention Pruning Strategy for Language Models (Li et al., EMNLP 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.emnlp-main.79.pdf
Video:
 https://aclanthology.org/2023.emnlp-main.79.mp4