Md Athikul Islam

Enhancing Adversarial Robustness in Natural Language Processing (NLP) Tasks Using Language Models

The adversarial robustness of natural language processing (NLP) models is critical for ensuring their reliability and trustworthiness across real-world applications. This dissertation investigates methods to enhance the robustness of NLP systems against adversarial attacks in diverse contexts, including general NLP tas...