arXiv paper: Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits
A new arXiv AI paper by Elena Dumitrescu, Gert Lek, and Lydia Y. Chen, and 1 more studies Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits.
ResearchAI
Follow arXiv AI/ML to make it a durable For You signal.