@haxorMB to Hacker NewsEnglish • 11 months agoUniversal and Transferable Adversarial Attacks on Aligned Language Modelsllm-attacks.orgmessage-square0fedilinkarrow-up13arrow-down11file-text
arrow-up12arrow-down1external-linkUniversal and Transferable Adversarial Attacks on Aligned Language Modelsllm-attacks.org@haxorMB to Hacker NewsEnglish • 11 months agomessage-square0fedilinkfile-text