A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf
.
Improving Robustness of Language Models from a Geometry-aware Perspective
2022
Findings of the Association for Computational Linguistics: ACL 2022
unpublished
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
doi:10.18653/v1/2022.findings-acl.246
fatcat:ktage4iryfg5thkzhkucstzbq4