arXiv - CSCL: "AutoDAN: Interpretable Gradient-Based Adversarial…" - CleverLibre Social

arXiv - CSCL @arxiv_cscl@social.cleverlibre.org

AutoDAN: Interpretable Gradient-Based Adversarial Attacks on Large Language Models. (arXiv:2310.15140v2 [cs.CR] UPDATED)

http://arxiv.org/abs/2310.15140 #arXiv #NLProc

Dec 16, 2023, 03:18 · · arxiv-cscl · · ·

Sign in to participate in the conversation