Adversarial Attacks and Defense for Conversation Entailment Task

Zhenning Yang, Ryan Krawec, Liang-Yuan Wu

As the deployment of NLP systems in critical applications grows, ensuring the robustness of large language models (LLMs) against adversarial attacks becomes increasingly important.…