all InfoSec news
Adversarial Nibbler: A Data-Centric Challenge for Improving the Safety of Text-to-Image Models. (arXiv:2305.14384v1 [cs.LG])
cs.CR updates on arXiv.org arxiv.org
The generative AI revolution in recent years has been spurred by an expansion
in compute power and data quantity, which together enable extensive
pre-training of powerful text-to-image (T2I) models. With their greater
capabilities to generate realistic and creative content, these T2I models like
DALL-E, MidJourney, Imagen or Stable Diffusion are reaching ever wider
audiences. Any unsafe behaviors inherited from pretraining on uncurated
internet-scraped datasets thus have the potential to cause wide-reaching harm,
for example, through generated images which are violent, …
adversarial capabilities challenge compute data enable generative generative ai midjourney power safety text training