notesum.ai

Published at December 10

PrisonBreak: Jailbreaking Large Language Models with Fewer Than Twenty-Five Targeted Bit-flips

cs.CR
cs.CL
cs.LG

Released Date: December 10, 2024

Authors: Zachary Coalson, Jeonghyun Woo, Shiyang Chen, Yu Sun, Lishan Yang, Prashant Nair, Bo Fang, Sanghyun Hong

Arxiv: http://arxiv.org/pdf/2412.07192v1