Fake authoritative citations matched to the type of harmful request can bypass safety alignment in several commercial and open LLMs.
”Superintelligence: Paths, dangers, strategies.” (2016): 196-203
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The Dark Side of Trust: Authority Citation-Driven Jailbreak Attacks on Large Language Models
Fake authoritative citations matched to the type of harmful request can bypass safety alignment in several commercial and open LLMs.