SemRoDe: Macro Adversarial Training to Learn Representations That are Robust to Word-Level Attacks

arXiv:2403.18423v1 Announce Type: cross
Abstract: Language models (LMs) are indispensable tools for natural language processing tasks, but their vulnerability to adversarial attacks remains a concern. While current research has explored adversarial training techniques, their improvements to defend against word-level attacks have been limited. In this work, we propose a novel approach called Semantic Robust Defence (SemRoDe), a Macro Adversarial Training strategy to enhance the robustness of LMs. Drawing inspiration from recent studies in the image domain, we investigate and later …

abstract adversarial adversarial attacks adversarial training arxiv attacks cs.cl cs.lg current improvements language language models language processing learn lms macro natural natural language natural language processing processing research robust tasks tools training type vulnerability word work