Emoti-Attack: How emojis can trick AI language models

Emoti-Attack: How emojis can trick AI language models

Deep neural networks (DNNs) have driven remarkable advancements in natural language processing (NLP), powering applications such as ChatGPT and automated content moderation systems. However, the vulnerability of these models to adversarial attacks remains a pressing concern. Unlike images, where slight modifications are often imperceptible, text operates in a discrete space, making even small alterations noticeable to human readers. This present…