Improving Robustness of Language Models from a Geometry-aware Perspective

Bin Zhu, Zhaoquan Gu, Le Wang, Jinyin Chen, Qi Xuan


Abstract
Recent studies have found that removing the norm-bounded projection and increasing search steps in adversarial training can significantly improve robustness. However, we observe that a too large number of search steps can hurt accuracy. We aim to obtain strong robustness efficiently using fewer steps. Through a toy experiment, we find that perturbing the clean data to the decision boundary but not crossing it does not degrade the test accuracy. Inspired by this, we propose friendly adversarial data augmentation (FADA) to generate friendly adversarial data. On top of FADA, we propose geometry-aware adversarial training (GAT) to perform adversarial training on friendly adversarial data so that we can save a large number of search steps. Comprehensive experiments across two widely used datasets and three pre-trained language models demonstrate that GAT can obtain stronger robustness via fewer steps. In addition, we provide extensive empirical results and in-depth analyses on robustness to facilitate future studies.
Anthology ID:
2022.findings-acl.246
Volume:
Findings of the Association for Computational Linguistics: ACL 2022
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3115–3125
Language:
URL:
https://aclanthology.org/2022.findings-acl.246
DOI:
10.18653/v1/2022.findings-acl.246
Bibkey:
Cite (ACL):
Bin Zhu, Zhaoquan Gu, Le Wang, Jinyin Chen, and Qi Xuan. 2022. Improving Robustness of Language Models from a Geometry-aware Perspective. In Findings of the Association for Computational Linguistics: ACL 2022, pages 3115–3125, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Improving Robustness of Language Models from a Geometry-aware Perspective (Zhu et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-acl.246.pdf
Software:
 2022.findings-acl.246.software.zip
Data
GLUEIMDb Movie ReviewsSSTSST-2