$\textit{LinkPrompt}$: Natural and Universal Adversarial Attacks on Prompt-based Language Models

arXiv:2403.16432v1 Announce Type: new
Abstract: Prompt-based learning is a new language model training paradigm that adapts the Pre-trained Language Models (PLMs) to downstream tasks, which revitalizes the performance benchmarks across various natural language processing (NLP) tasks. Instead of using a fixed prompt template to fine-tune the model, some research demonstrates the effectiveness of searching for the prompt via optimization. Such prompt optimization process of prompt-based learning on PLMs also gives insight into generating adversarial prompts to mislead the model, raising …

abstract adversarial adversarial attacks arxiv attacks benchmarks cs.ai cs.cl language language model language models language model training language processing natural natural language natural language processing nlp paradigm performance processing prompt prompt-based learning research tasks template training type universal