Ghostwriter attack injects fabricated evidence to steer LLM viewpoints, with experiments showing high success on commercial models and partial mitigation on guarded ones.
Can editing llms inject harm?
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CR 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
Survey of harmful fine-tuning attacks on LLMs, their variants, defense strategies, mechanical analysis, and evaluation methodologies.
citing papers explorer
-
Steering LLM Viewpoints through Fabricated Evidence Injection
Ghostwriter attack injects fabricated evidence to steer LLM viewpoints, with experiments showing high success on commercial models and partial mitigation on guarded ones.
-
Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
Survey of harmful fine-tuning attacks on LLMs, their variants, defense strategies, mechanical analysis, and evaluation methodologies.