GuardT2I: Defending Text-to-Image Models from Adversarial Prompts

arXiv:2403.01446v1 Announce Type: new
Abstract: Recent advancements in Text-to-Image (T2I) models have raised significant safety concerns about their potential misuse for generating inappropriate or Not-Safe-For-Work (NSFW) contents, despite existing countermeasures such as NSFW classifiers or model fine-tuning for inappropriate concept removal. Addressing this challenge, our study unveils GuardT2I, a novel moderation framework that adopts a generative approach to enhance T2I models' robustness against adversarial prompts. Instead of making a binary classification, GuardT2I utilizes a Large Language Model (LLM) to conditionally …

abstract adversarial arxiv challenge classifiers concept concerns contents cs.cv fine-tuning framework image inappropriate misuse model fine-tuning moderation novel nsfw prompts safety study text text-to-image type work